Release v0.6.0

docs(web): update social preview image
2026-08-09 17:17:57 +05:30 · 2026-07-01 15:15:08 +05:30 · 2026-07-01 15:12:47 +05:30 · 2026-07-01 14:51:27 +05:30 · 2026-07-01 14:42:10 +05:30 · 2026-06-29 17:44:10 +05:30
182 changed files with 38307 additions and 3022 deletions
@@ -0,0 +1,52 @@
+# Dependencies & Build artifacts
+target/
+**/target/
+
+# Environment & secrets
+.env
+.env.*
+*.key
+*.pem
+
+# Development & testing files
+.git/
+.gitignore
+.github/
+.gitattributes
+
+# Documentation & notes
+README*.md
+docs/
+CONTRIBUTING.md
+CHANGELOG.md
+LICENSE
+
+# Logs & temporary files
+*.log
+*.tmp
+data/
+backups/
+
+# Editor & IDE files
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# Docker related
+Dockerfile*
+.dockerignore
+docker-compose*.yml
+.docker/
+
+# Test files
+tests/
+__tests__/
+*.test.*
+*.spec.*
+
+# Other unnecessary files
+node_modules/
+dist/
+build/
@@ -1,22 +1,85 @@
-name: Rust
+  name: Rust CI

-on:
-  push:
-    branches: [ "main" ]
-  pull_request:
-    branches: [ "main" ]
+  on:
+    push:
+      branches: ["main"]
+    pull_request:
+      branches: ["main"]

-env:
-  CARGO_TERM_COLOR: always
+  env:
+    CARGO_TERM_COLOR: always
+    CARGO_INCREMENTAL: 0
+    REGISTRY: ghcr.io
+    IMAGE_NAME: ${{ github.repository }}

-jobs:
-  build:
+  jobs:
+    test:
+      name: Test & Quality Checks
+      runs-on: ubuntu-latest

-    runs-on: ubuntu-latest
+      steps:
+        - name: Checkout Repository
+          uses: actions/checkout@v4

-    steps:
-    - uses: actions/checkout@v4
-    - name: Build
-      run: cargo build --verbose
-    - name: Run tests
-      run: cargo test --verbose
+        - name: Install Rust Toolchain
+          uses: dtolnay/rust-toolchain@stable
+          with:
+            components: rustfmt, clippy
+
+        - name: Cache Cargo Dependencies
+          uses: Swatinem/rust-cache@v2
+
+        - name: Check Formatting
+          run: cargo fmt --check
+
+        - name: Run Clippy
+          run: cargo clippy --all-targets --all-features -- -D warnings
+
+        - name: Build Release
+          run: cargo build --release --verbose
+
+        - name: Run Tests
+          run: cargo test --all-features --verbose
+
+    docker:
+      name: Build Docker Image
+      runs-on: ubuntu-latest
+      needs: test
+
+      permissions:
+        contents: read
+        packages: write
+
+      steps:
+        - name: Checkout Repository
+          uses: actions/checkout@v4
+
+        - name: Login to GitHub Container Registry
+          uses: docker/login-action@v3
+          with:
+            registry: ghcr.io
+            username: ${{ github.actor }}
+            password: ${{ secrets.GITHUB_TOKEN }}
+
+        - name: Extract Docker Metadata
+          id: meta
+          uses: docker/metadata-action@v5
+          with:
+            images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+            tags: |
+              type=sha
+              type=raw,value=latest,enable={{is_default_branch}}
+
+        - name: Setup Docker Buildx
+          uses: docker/setup-buildx-action@v3
+
+        - name: Build and Push Docker Image
+          uses: docker/build-push-action@v6
+          with:
+            context: .
+            file: ./Dockerfile
+            push: ${{ github.ref == 'refs/heads/main' }}
+            tags: ${{ steps.meta.outputs.tags }}
+            labels: ${{ steps.meta.outputs.labels }}
+            cache-from: type=gha
+            cache-to: type=gha,mode=max
@@ -1,4 +1,4 @@
-/target
+/target/
 data/
 *.db
 *.db-wal
@@ -29,24 +29,6 @@ dependencies = [
 "memchr",
 ]

-[[package]]
-name = "aligned"
-version = "0.4.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ee4508988c62edf04abd8d92897fca0c2995d907ce1dfeaf369dac3716a40685"
-dependencies = [
- "as-slice",
-]
-
-[[package]]
-name = "aligned-vec"
-version = "0.6.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dc890384c8602f339876ded803c97ad529f3842aba97f6392b3dba0dd171769b"
-dependencies = [
- "equator",
-]
-
 [[package]]
 name = "android_system_properties"
 version = "0.1.5"
@@ -121,17 +103,6 @@ dependencies = [
 "derive_arbitrary",
 ]

-[[package]]
-name = "arg_enum_proc_macro"
-version = "0.3.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0ae92a5119aa49cdbcf6b9f893fe4e1d98b04ccbf82ee0584ad948a44a734dea"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn",
-]
-
 [[package]]
 name = "argon2"
 version = "0.5.3"
@@ -144,21 +115,6 @@ dependencies = [
 "password-hash",
 ]

-[[package]]
-name = "arrayvec"
-version = "0.7.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7c02d123df017efcdfbd739ef81735b36c5ba83ec3c59c80a9d7ecc718f92e50"
-
-[[package]]
-name = "as-slice"
-version = "0.2.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "516b6b4f0e40d50dcda9365d53964ec74560ad4284da2e7fc97122cd83174516"
-dependencies = [
- "stable_deref_trait",
-]
-
 [[package]]
 name = "askama"
 version = "0.12.1"
@@ -200,7 +156,7 @@ version = "0.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "acb1161c6b64d1c3d83108213c2a2533a342ac225aabd0bda218278c2ddb00c0"
 dependencies = [
- "nom 7.1.3",
+ "nom",
 ]

 [[package]]
@@ -226,49 +182,6 @@ version = "1.5.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f2032f911046de80f0a198e0901378627c33f59ea0ac00e363d481118bd70a53"

-[[package]]
-name = "av-scenechange"
-version = "0.14.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0f321d77c20e19b92c39e7471cf986812cbb46659d2af674adc4331ef3f18394"
-dependencies = [
- "aligned",
- "anyhow",
- "arg_enum_proc_macro",
- "arrayvec",
- "log",
- "num-rational",
- "num-traits",
- "pastey",
- "rayon",
- "thiserror",
- "v_frame",
- "y4m",
-]
-
-[[package]]
-name = "av1-grain"
-version = "0.2.5"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8cfddb07216410377231960af4fcab838eaa12e013417781b78bd95ee22077f8"
-dependencies = [
- "anyhow",
- "arrayvec",
- "log",
- "nom 8.0.0",
- "num-rational",
- "v_frame",
-]
-
-[[package]]
-name = "avif-serialize"
-version = "0.8.9"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e7178fe5f7d460b13895ebb9dcb28a3a6216d2df2574a0806cb51b555d297f38"
-dependencies = [
- "arrayvec",
-]
-
 [[package]]
 name = "axum"
 version = "0.7.9"
@@ -289,6 +202,7 @@ dependencies = [
 "matchit",
 "memchr",
 "mime",
+ "multer",
 "percent-encoding",
 "pin-project-lite",
 "rustversion",
@@ -381,27 +295,12 @@ dependencies = [
 "serde",
 ]

-[[package]]
-name = "bit_field"
-version = "0.10.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1e4b40c7323adcfc0a41c4b88143ed58346ff65a288fc144329c5c45e05d70c6"
-
 [[package]]
 name = "bitflags"
 version = "2.13.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b4388bee8683e3d04af747c73422af53102d2bd24d9eadb6cbc100baef4b43f8"

-[[package]]
-name = "bitstream-io"
-version = "4.10.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7eff00be299a18769011411c9def0d827e8f2d7bf0c3dbf53633147a8867fd1f"
-dependencies = [
- "no_std_io2",
-]
-
 [[package]]
 name = "blake2"
 version = "0.10.6"
@@ -420,12 +319,6 @@ dependencies = [
 "generic-array",
 ]

-[[package]]
-name = "built"
-version = "0.8.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5c0e531d93d39c34eef561e929e8a7f86d77a5af08aac4f6d6e39976c51858e9"
-
 [[package]]
 name = "bumpalo"
 version = "3.20.3"
@@ -452,7 +345,7 @@ checksum = "1e748733b7cbc798e1434b6ac524f0c1ff2ab456fe201501e6497c8417a4fc33"

 [[package]]
 name = "bzod"
-version = "0.1.0"
+version = "0.6.0"
 dependencies = [
 "argon2",
 "askama",
@@ -462,6 +355,7 @@ dependencies = [
 "clap",
 "dotenvy",
 "flate2",
+ "futures-util",
 "hex",
 "image",
 "qrcode",
@@ -479,6 +373,7 @@ dependencies = [
 "tracing-subscriber",
 "uuid",
 "zip",
+ "zstd",
 ]

 [[package]]
@@ -559,12 +454,6 @@ version = "1.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c8d4a3bb8b1e0c1050499d1815f5ab16d04f0959b233085fb31653fbfc9d98f9"

-[[package]]
-name = "color_quant"
-version = "1.1.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3d7b894f5411737b7867f4827955924d7c254fc9f4d91a6aad6b097804b1018b"
-
 [[package]]
 name = "colorchoice"
 version = "1.0.5"
@@ -582,6 +471,24 @@ dependencies = [
 "version_check",
 ]

+[[package]]
+name = "cookie_store"
+version = "0.22.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "15b2c103cf610ec6cae3da84a766285b42fd16aad564758459e6ecf128c75206"
+dependencies = [
+ "cookie",
+ "document-features",
+ "idna",
+ "log",
+ "publicsuffix",
+ "serde",
+ "serde_derive",
+ "serde_json",
+ "time",
+ "url",
+]
+
 [[package]]
 name = "core-foundation-sys"
 version = "0.8.7"
@@ -606,37 +513,12 @@ dependencies = [
 "cfg-if",
 ]

-[[package]]
-name = "crossbeam-deque"
-version = "0.8.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9dd111b7b7f7d55b72c0a6ae361660ee5853c9af73f70c3c2ef6858b950e2e51"
-dependencies = [
- "crossbeam-epoch",
- "crossbeam-utils",
-]
-
-[[package]]
-name = "crossbeam-epoch"
-version = "0.9.18"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5b82ac4a3c2ca9c3460964f020e1402edd5753411d7737aa39c3714ad1b5420e"
-dependencies = [
- "crossbeam-utils",
-]
-
 [[package]]
 name = "crossbeam-utils"
 version = "0.8.21"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d0a5c400df2834b80a4c3327b3aad3a4c4cd4de0629063962b03235697506a28"

-[[package]]
-name = "crunchy"
-version = "0.2.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "460fbee9c2c2f33933d720630a6a0bac33ba7053db5344fac858d4b8952d77d5"
-
 [[package]]
 name = "crypto-common"
 version = "0.1.7"
@@ -689,18 +571,21 @@ dependencies = [
 "syn",
 ]

+[[package]]
+name = "document-features"
+version = "0.2.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d4b8a88685455ed29a21542a33abd9cb6510b6b129abadabdcef0f4c55bc8f61"
+dependencies = [
+ "litrs",
+]
+
 [[package]]
 name = "dotenvy"
 version = "0.15.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1aaf95b3e5c8f23aa320147307562d361db0ae0d51242340f558153b4eb2439b"

-[[package]]
-name = "either"
-version = "1.16.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "91622ff5e7162018101f2fea40d6ebf4a78bbe5a49736a2020649edf9693679e"
-
 [[package]]
 name = "encoding_rs"
 version = "0.8.35"
@@ -710,26 +595,6 @@ dependencies = [
 "cfg-if",
 ]

-[[package]]
-name = "equator"
-version = "0.4.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4711b213838dfee0117e3be6ac926007d7f433d7bbe33595975d4190cb07e6fc"
-dependencies = [
- "equator-macro",
-]
-
-[[package]]
-name = "equator-macro"
-version = "0.4.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "44f23cf4b44bfce11a86ace86f8a73ffdec849c9fd00a386a53d278bd9e81fb3"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn",
-]
-
 [[package]]
 name = "equivalent"
 version = "1.0.2"
@@ -746,21 +611,6 @@ dependencies = [
 "windows-sys 0.61.2",
 ]

-[[package]]
-name = "exr"
-version = "1.74.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4300e043a56aa2cb633c01af81ca8f699a321879a7854d3896a0ba89056363be"
-dependencies = [
- "bit_field",
- "half",
- "lebe",
- "miniz_oxide",
- "rayon-core",
- "smallvec",
- "zune-inflate",
-]
-
 [[package]]
 name = "fallible-iterator"
 version = "0.3.0"
@@ -779,12 +629,6 @@ version = "2.4.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9f1f227452a390804cdb637b74a86990f2a7d7ba4b7d5693aac9b4dd6defd8d6"

-[[package]]
-name = "fax"
-version = "0.2.7"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "caf1079563223d5d59d83c85886a56e586cfd5c1a26292e971a0fa266531ac5a"
-
 [[package]]
 name = "fdeflate"
 version = "0.3.7"
@@ -850,6 +694,17 @@ version = "0.3.32"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7e3450815272ef58cec6d564423f6e755e25379b217b0bc688e295ba24df6b1d"

+[[package]]
+name = "futures-macro"
+version = "0.3.32"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e835b70203e41293343137df5c0664546da5745f82ec9b84d40be8336958447b"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "futures-task"
 version = "0.3.32"
@@ -863,6 +718,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "389ca41296e6190b48053de0321d02a77f32f8a5d2461dd38762c0593805c6d6"
 dependencies = [
 "futures-core",
+ "futures-macro",
 "futures-task",
 "pin-project-lite",
 "slab",
@@ -918,27 +774,6 @@ dependencies = [
 "wasip3",
 ]

-[[package]]
-name = "gif"
-version = "0.14.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ee8cfcc411d9adbbaba82fb72661cc1bcca13e8bba98b364e62b2dba8f960159"
-dependencies = [
- "color_quant",
- "weezl",
-]
-
-[[package]]
-name = "half"
-version = "2.7.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6ea2d84b969582b4b1864a92dc5d27cd2b77b622a8d79306834f1be5ba20d84b"
-dependencies = [
- "cfg-if",
- "crunchy",
- "zerocopy",
-]
-
 [[package]]
 name = "hashbrown"
 version = "0.14.5"
@@ -1239,38 +1074,11 @@ checksum = "85ab80394333c02fe689eaf900ab500fbd0c2213da414687ebf995a65d5a6104"
 dependencies = [
 "bytemuck",
 "byteorder-lite",
- "color_quant",
- "exr",
- "gif",
- "image-webp",
 "moxcms",
 "num-traits",
 "png",
- "qoi",
- "ravif",
- "rayon",
- "rgb",
- "tiff",
- "zune-core",
- "zune-jpeg",
 ]

-[[package]]
-name = "image-webp"
-version = "0.2.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "525e9ff3e1a4be2fbea1fdf0e98686a6d98b4d8f937e1bf7402245af1909e8c3"
-dependencies = [
- "byteorder-lite",
- "quick-error",
-]
-
-[[package]]
-name = "imgref"
-version = "1.12.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "89194689a993ab15268672e99e7b0e19da2da3268ac682e8f02d29d4d1434cd7"
-
 [[package]]
 name = "indexmap"
 version = "2.14.0"
@@ -1283,17 +1091,6 @@ dependencies = [
 "serde_core",
 ]

-[[package]]
-name = "interpolate_name"
-version = "0.2.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c34819042dc3d3971c46c2190835914dfbe0c3c13f61449b2997f4e9722dfa60"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn",
-]
-
 [[package]]
 name = "ipnet"
 version = "2.12.0"
@@ -1306,15 +1103,6 @@ version = "1.70.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a6cb138bb79a146c1bd460005623e142ef0181e3d0219cb493e02f7d08a35695"

-[[package]]
-name = "itertools"
-version = "0.14.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2b192c782037fadd9cfa75548310488aabdbf3d2da73885b31bd0abd03351285"
-dependencies = [
- "either",
-]
-
 [[package]]
 name = "itoa"
 version = "1.0.18"
@@ -1354,28 +1142,12 @@ version = "0.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "09edd9e8b54e49e587e4f6295a7d29c3ea94d469cb40ab8ca70b288248a81db2"

-[[package]]
-name = "lebe"
-version = "0.5.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7a79a3332a6609480d7d0c9eab957bca6b455b91bb84e66d19f5ff66294b85b8"
-
 [[package]]
 name = "libc"
 version = "0.2.186"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "68ab91017fe16c622486840e4c83c9a37afeff978bd239b5293d61ece587de66"

-[[package]]
-name = "libfuzzer-sys"
-version = "0.4.13"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a9fd2f41a1cba099f79a0b6b6c35656cf7c03351a7bae8ff0f28f25270f929d2"
-dependencies = [
- "arbitrary",
- "cc",
-]
-
 [[package]]
 name = "libm"
 version = "0.2.16"
@@ -1405,6 +1177,12 @@ version = "0.8.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "92daf443525c4cce67b150400bc2316076100ce0b3686209eb8cf3c31612e6f0"

+[[package]]
+name = "litrs"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "11d3d7f243d5c5a8b9bb5d6dd2b1602c0cb0b9db1621bafc7ed66e35ff9fe092"
+
 [[package]]
 name = "lock_api"
 version = "0.4.14"
@@ -1420,15 +1198,6 @@ version = "0.4.32"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "953f07c43838f8e6f9758cab68bf5bed85465e7587ebe0b823f1bcd81978ad3a"

-[[package]]
-name = "loop9"
-version = "0.1.5"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0fae87c125b03c1d2c0150c90365d7d6bcc53fb73a9acaef207d2d065860f062"
-dependencies = [
- "imgref",
-]
-
 [[package]]
 name = "lru-slab"
 version = "0.1.2"
@@ -1450,16 +1219,6 @@ version = "0.7.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0e7465ac9959cc2b1404e8e2367b43684a6d13790fe23056cc8c6c5a6b7bcb94"

-[[package]]
-name = "maybe-rayon"
-version = "0.1.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8ea1f30cedd69f0a2954655f7188c6a834246d2bcf1e315e2ac40c4b24dc9519"
-dependencies = [
- "cfg-if",
- "rayon",
-]
-
 [[package]]
 name = "memchr"
 version = "2.8.1"
@@ -1536,21 +1295,6 @@ dependencies = [
 "version_check",
 ]

-[[package]]
-name = "new_debug_unreachable"
-version = "1.0.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "650eef8c711430f1a879fdd01d4745a7deea475becfb90269c06775983bbf086"
-
-[[package]]
-name = "no_std_io2"
-version = "0.9.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "418abd1b6d34fbf6cae440dc874771b0525a604428704c76e48b29a5e67b8003"
-dependencies = [
- "memchr",
-]
-
 [[package]]
 name = "nom"
 version = "7.1.3"
@@ -1561,21 +1305,6 @@ dependencies = [
 "minimal-lexical",
 ]

-[[package]]
-name = "nom"
-version = "8.0.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "df9761775871bdef83bee530e60050f7e54b1105350d6884eb0fb4f46c2f9405"
-dependencies = [
- "memchr",
-]
-
-[[package]]
-name = "noop_proc_macro"
-version = "0.3.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0676bb32a98c1a483ce53e500a81ad9c3d5b3f7c920c28c24e9cb0980d0b5bc8"
-
 [[package]]
 name = "nu-ansi-term"
 version = "0.50.3"
@@ -1585,53 +1314,12 @@ dependencies = [
 "windows-sys 0.61.2",
 ]

-[[package]]
-name = "num-bigint"
-version = "0.4.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a5e44f723f1133c9deac646763579fdb3ac745e418f2a7af9cd0c431da1f20b9"
-dependencies = [
- "num-integer",
- "num-traits",
-]
-
 [[package]]
 name = "num-conv"
 version = "0.2.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "521739c6d2bac4aa25192232afe6841231376b2b26d4d9fae5ecf8ca5772e441"

-[[package]]
-name = "num-derive"
-version = "0.4.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ed3955f1a9c7c0c15e092f9c887db08b1fc683305fdf6eb6684f22555355e202"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn",
-]
-
-[[package]]
-name = "num-integer"
-version = "0.1.46"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7969661fd2958a5cb096e56c8e1ad0444ac2bbcd0061bd28660485a44879858f"
-dependencies = [
- "num-traits",
-]
-
-[[package]]
-name = "num-rational"
-version = "0.4.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f83d14da390562dca69fc84082e73e548e1ad308d24accdedd2720017cb37824"
-dependencies = [
- "num-bigint",
- "num-integer",
- "num-traits",
-]
-
 [[package]]
 name = "num-traits"
 version = "0.2.19"
@@ -1687,18 +1375,6 @@ dependencies = [
 "subtle",
 ]

-[[package]]
-name = "paste"
-version = "1.0.15"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "57c0d7b74b563b49d38dae00a0c37d4d6de9b432382b2892f0574ddcae73fd0a"
-
-[[package]]
-name = "pastey"
-version = "0.1.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "35fb2e5f958ec131621fdd531e9fc186ed768cbe395337403ae56c17a74c68ec"
-
 [[package]]
 name = "percent-encoding"
 version = "2.3.2"
@@ -1774,22 +1450,19 @@ dependencies = [
 ]

 [[package]]
-name = "profiling"
-version = "1.0.18"
+name = "psl-types"
+version = "2.0.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3d595e54a326bc53c1c197b32d295e14b169e3cfeaa8dc82b529f947fba6bcf5"
-dependencies = [
- "profiling-procmacros",
-]
+checksum = "33cb294fe86a74cbcf50d4445b37da762029549ebeea341421c7c70370f86cac"

 [[package]]
-name = "profiling-procmacros"
-version = "1.0.18"
+name = "publicsuffix"
+version = "2.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4488a4a36b9a4ba6b9334a32a39971f77c1436ec82c38707bce707699cc3bbcb"
+checksum = "6f42ea446cab60335f76979ec15e12619a2165b5ae2c12166bef27d283a9fadf"
 dependencies = [
- "quote",
- "syn",
+ "idna",
+ "psl-types",
 ]

 [[package]]
@@ -1798,15 +1471,6 @@ version = "0.1.29"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e0c5ccf5294c6ccd63a74f1565028353830a9c2f5eb0c682c355c471726a6e3f"

-[[package]]
-name = "qoi"
-version = "0.4.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7f6d64c71eb498fe9eae14ce4ec935c555749aef511cca85b5568910d6e48001"
-dependencies = [
- "bytemuck",
-]
-
 [[package]]
 name = "qrcode"
 version = "0.14.1"
@@ -1816,12 +1480,6 @@ dependencies = [
 "image",
 ]

-[[package]]
-name = "quick-error"
-version = "2.0.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a993555f31e5a609f617c12db6250dedcac1b0a85076912c436e6fc9b2c8e6a3"
-
 [[package]]
 name = "quinn"
 version = "0.11.9"
@@ -1957,76 +1615,6 @@ dependencies = [
 "getrandom 0.3.4",
 ]

-[[package]]
-name = "rav1e"
-version = "0.8.1"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "43b6dd56e85d9483277cde964fd1bdb0428de4fec5ebba7540995639a21cb32b"
-dependencies = [
- "aligned-vec",
- "arbitrary",
- "arg_enum_proc_macro",
- "arrayvec",
- "av-scenechange",
- "av1-grain",
- "bitstream-io",
- "built",
- "cfg-if",
- "interpolate_name",
- "itertools",
- "libc",
- "libfuzzer-sys",
- "log",
- "maybe-rayon",
- "new_debug_unreachable",
- "noop_proc_macro",
- "num-derive",
- "num-traits",
- "paste",
- "profiling",
- "rand 0.9.4",
- "rand_chacha 0.9.0",
- "simd_helpers",
- "thiserror",
- "v_frame",
- "wasm-bindgen",
-]
-
-[[package]]
-name = "ravif"
-version = "0.13.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e52310197d971b0f5be7fe6b57530dcd27beb35c1b013f29d66c1ad73fbbcc45"
-dependencies = [
- "avif-serialize",
- "imgref",
- "loop9",
- "quick-error",
- "rav1e",
- "rayon",
- "rgb",
-]
-
-[[package]]
-name = "rayon"
-version = "1.12.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fb39b166781f92d482534ef4b4b1b2568f42613b53e5b6c160e24cfbfa30926d"
-dependencies = [
- "either",
- "rayon-core",
-]
-
-[[package]]
-name = "rayon-core"
-version = "1.13.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "22e18b0f0062d30d4230b2e85ff77fdfe4326feb054b9783a3460d8435c8ab91"
-dependencies = [
- "crossbeam-deque",
- "crossbeam-utils",
-]
-
 [[package]]
 name = "redox_syscall"
 version = "0.5.18"
@@ -2061,6 +1649,8 @@ checksum = "eddd3ca559203180a307f12d114c268abf583f59b03cb906fd0b3ff8646c1147"
 dependencies = [
 "base64",
 "bytes",
+ "cookie",
+ "cookie_store",
 "futures-core",
 "http",
 "http-body",
@@ -2091,12 +1681,6 @@ dependencies = [
 "webpki-roots",
 ]

-[[package]]
-name = "rgb"
-version = "0.8.53"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "47b34b781b31e5d73e9fbc8689c70551fd1ade9a19e3e28cfec8580a79290cc4"
-
 [[package]]
 name = "ring"
 version = "0.17.14"
@@ -2320,15 +1904,6 @@ version = "0.3.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "703d5c7ef118737c72f1af64ad2f6f8c5e1921f818cdcb97b8fe6fc69bf66214"

-[[package]]
-name = "simd_helpers"
-version = "0.1.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "95890f873bec569a0362c235787f3aca6e1e887302ba4840839bcc6459c42da6"
-dependencies = [
- "quote",
-]
-
 [[package]]
 name = "slab"
 version = "0.4.12"
@@ -2446,20 +2021,6 @@ dependencies = [
 "cfg-if",
 ]

-[[package]]
-name = "tiff"
-version = "0.11.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b63feaf3343d35b6ca4d50483f94843803b0f51634937cc2ec519fc32232bc52"
-dependencies = [
- "fax",
- "flate2",
- "half",
- "quick-error",
- "weezl",
- "zune-jpeg",
-]
-
 [[package]]
 name = "time"
 version = "0.3.47"
@@ -2775,17 +2336,6 @@ dependencies = [
 "wasm-bindgen",
 ]

-[[package]]
-name = "v_frame"
-version = "0.3.9"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "666b7727c8875d6ab5db9533418d7c764233ac9c0cff1d469aec8fa127597be2"
-dependencies = [
- "aligned-vec",
- "num-traits",
- "wasm-bindgen",
-]
-
 [[package]]
 name = "valuable"
 version = "0.1.1"
@@ -2955,12 +2505,6 @@ dependencies = [
 "rustls-pki-types",
 ]

-[[package]]
-name = "weezl"
-version = "0.1.12"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a28ac98ddc8b9274cb41bb4d9d4d5c425b6020c50c46f25559911905610b4a88"
-
 [[package]]
 name = "windows-core"
 version = "0.62.2"
@@ -3295,12 +2839,6 @@ dependencies = [
 "rustix",
 ]

-[[package]]
-name = "y4m"
-version = "0.8.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7a5a4b21e1a62b67a2970e6831bc091d7b87e119e7f9791aef9702e3bef04448"
-
 [[package]]
 name = "yoke"
 version = "0.8.3"
@@ -3440,25 +2978,29 @@ dependencies = [
 ]

 [[package]]
-name = "zune-core"
-version = "0.5.1"
+name = "zstd"
+version = "0.13.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "cb8a0807f7c01457d0379ba880ba6322660448ddebc890ce29bb64da71fb40f9"
-
-[[package]]
-name = "zune-inflate"
-version = "0.2.54"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "73ab332fe2f6680068f3582b16a24f90ad7096d5d39b974d1c0aff0125116f02"
+checksum = "e91ee311a569c327171651566e07972200e76fcfe2242a4fa446149a3881c08a"
 dependencies = [
- "simd-adler32",
+ "zstd-safe",
 ]

 [[package]]
-name = "zune-jpeg"
-version = "0.5.15"
+name = "zstd-safe"
+version = "7.2.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "27bc9d5b815bc103f142aa054f561d9187d191692ec7c2d1e2b4737f8dbd7296"
+checksum = "8f49c4d5f0abb602a93fb8736af2a4f4dd9512e36f7f570d66e65ff867ed3b9d"
 dependencies = [
- "zune-core",
+ "zstd-sys",
+]
+
+[[package]]
+name = "zstd-sys"
+version = "2.0.16+zstd.1.5.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "91e19ebc2adc8f83e43039e79776e3fda8ca919132d68a1fed6a5faca2683748"
+dependencies = [
+ "cc",
+ "pkg-config",
 ]
@@ -1,11 +1,31 @@
 [package]
 name = "bzod"
-version = "0.1.0"
+description = "Self-hosted multi-user URL management, landing page and QR analytics platform"
+version = "0.6.0"
 edition = "2021"
+license = "MIT OR Apache-2.0"
+repository = "https://github.com/thakares/nx9-url-shortener"
+homepage = "https://bzo.in"
+documentation = "https://github.com/thakares/nx9-url-shortener"
+readme = "README.md"
+authors = ["Sunil P. Thakare"]
+
+keywords = [
+    "url-shortener",
+    "landing-pages",
+    "analytics",
+    "qr-code",
+    "self-hosted"
+]
+
+categories = [
+    "web-programming",
+    "command-line-utilities"
+]

 [dependencies]
 tokio = { version = "1", features = ["full"] }
-axum = { version = "0.7", features = ["macros"] }
+axum = { version = "0.7", features = ["macros", "multipart"] }
 axum-extra = { version = "0.9", features = ["cookie"] }
 rusqlite = { version = "0.31", features = ["bundled"] }
 serde = { version = "1.0", features = ["derive"] }
@@ -19,7 +39,7 @@ askama = { version = "0.12" }
 argon2 = "0.5"
 sha2 = "0.10"
 rand = "0.8"
-reqwest = { version = "0.12", default-features = false, features = ["rustls-tls", "json"] }
+reqwest = { version = "0.12", default-features = false, features = ["rustls-tls", "json", "cookies"] }
 tar = "0.4"
 flate2 = "1.0"
 chrono = { version = "0.4", features = ["serde"] }
@@ -27,5 +47,17 @@ hex = "0.4"
 time = "0.3"
 toml = "0.8"
 qrcode = "0.14"
-image = "0.25"
+image = { version = "0.25", default-features = false, features = ["png"] }
 zip = { version = "2.1", default-features = false, features = ["deflate"] }
+futures-util = "0.3"
+zstd = "0.13"
+
+[lints.clippy]
+let_unit_value = "allow"
+useless_vec = "allow"
+
+[profile.release]
+lto = true
+codegen-units = 1
+strip = true
+panic = "abort"
@@ -1,7 +1,6 @@
 # ==========================================
-# Stage 1: Build
+# Stage 1: Builder (with optimized caching)
 # ==========================================
-# FROM rust:1.82-slim-bookworm AS builder
 FROM rust:1.89-bookworm AS builder

 WORKDIR /app
@@ -10,34 +9,33 @@ WORKDIR /app
 RUN apt-get update && apt-get install -y \
    pkg-config \
    libssl-dev \
-    git \
    && rm -rf /var/lib/apt/lists/*

-# Copy configuration files
-# COPY Cargo.toml ./
+# Copy only Cargo files first (best caching)
 COPY Cargo.toml Cargo.lock ./

-# Pre-build dependencies to cache them
-RUN mkdir src && echo "fn main() {}" > src/main.rs
-RUN cargo build --release
-RUN rm -rf src
+# Create dummy source for dependency caching
+RUN mkdir -p src && \
+    echo "fn main() { println!(\"dummy\"); }" > src/main.rs && \
+    cargo build --release && \
+    rm -rf src target/release/deps/bzod*

-# Copy source and templates
+# Copy real source code + assets
 COPY src ./src
 COPY templates ./templates
+COPY www ./www

-# Trigger rebuilding with actual source
-RUN touch src/main.rs
+# Build the real application
 RUN cargo build --release

 # ==========================================
-# Stage 2: Runner
+# Stage 2: Runtime (slim)
 # ==========================================
 FROM debian:bookworm-slim

 WORKDIR /app

-# Install runtime dependencies
+# Runtime dependencies
 RUN apt-get update && apt-get install -y \
    openssl \
    ca-certificates \
@@ -47,23 +45,29 @@ RUN apt-get update && apt-get install -y \
 # Copy binary from builder
 COPY --from=builder /app/target/release/bzod /usr/local/bin/bzod

-# Create non-root user and data directory
+# Copy assets
+COPY --from=builder /app/templates ./templates
+COPY --from=builder /app/www ./www
+
+# Create non-root user
 RUN groupadd -g 1000 bzod && \
    useradd -u 1000 -g bzod -m -s /bin/bash bzod

-RUN mkdir -p /app/data && chown -R bzod:bzod /app/data
+# Create data directory
+RUN mkdir -p /app/data && \
+    chown -R bzod:bzod /app

 USER bzod

-ENV DATA_DIR=/app/data
-ENV PORT=8654
-ENV HOST=0.0.0.0
-ENV COOKIE_SECURE=true
+ENV DATA_DIR=/app/data \
+    PORT=8654 \
+    HOST=0.0.0.0 \
+    COOKIE_SECURE=true

 EXPOSE 8654

 HEALTHCHECK --interval=30s --timeout=5s --start-period=5s --retries=3 \
-  CMD curl -f http://localhost:$${PORT:-8654}/status || exit 1
+  CMD curl -f http://localhost:${PORT}/status || exit 1

 ENTRYPOINT ["bzod"]
-CMD ["serve"]
+CMD ["serve"]
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Sunil Purushottam Thakare
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
@@ -1,11 +1,11 @@
 #!/usr/bin/env bash
-
-# BZOD Deployment Script (Debian Native Deployment)
-# This script sets up a secure, production-ready systemd service for BZOD.
+# BZOD Production Deployment Script
+# curl -fsSL https://bzo.in/deploy.sh | sudo bash

 set -euo pipefail

-# Configurations
+BZOD_VERSION="0.6.0"
+
 SERVICE_USER="bzod"
 INSTALL_PATH="/usr/local/bin/bzod"
 CONFIG_DIR="/etc/bzod"
@@ -13,93 +13,145 @@ DATA_DIR="/var/lib/bzod/data"
 ENV_FILE="${CONFIG_DIR}/bzod.env"
 SYSTEMD_UNIT="/etc/systemd/system/bzod.service"

-# Color outputs
 RED='\033[0;31m'
 GREEN='\033[0;32m'
 BLUE='\033[0;34m'
-NC='\033[0m' # No Color
+NC='\033[0m'

-echo -e "${BLUE}=== BZOD Debian Deployment Script ===${NC}"
+# Temporary file cleanup
+TMP_BINARY=""
+cleanup() {
+    rm -f "${TMP_BINARY:-}" "${TMP_GHCR:-}"
+}
+trap cleanup EXIT
+
+echo -e "${BLUE}=== BZOD - Privacy-First URL Shortener & Landing Page Platform ===${NC}"
+echo -e "Production deployment started...\n"

-# 1. Check Root Privileges
 if [ "$EUID" -ne 0 ]; then
-    echo -e "${RED}Error: This script must be run as root (or via sudo).${NC}"
+    echo -e "${RED}Error: This script must be run as root (use sudo).${NC}"
    exit 1
 fi

-# 2. Install Package Dependencies
-echo -e "\n${BLUE}[1/8] Installing system dependencies (SQLite, OpenSSL, Tar)...${NC}"
-apt-get update
+# 1. Install Base Dependencies
+echo -e "${BLUE}[1/8] Installing base system dependencies...${NC}"
+apt-get update -qq
 apt-get install -y openssl sqlite3 ca-certificates curl tar gzip

-# 3. Compile Production Build Locally
-echo -e "\n${BLUE}[2/8] Compiling release binary...${NC}"
-if ! command -v cargo &> /dev/null; then
-    echo -e "${RED}Error: cargo not found. Please install Rust or copy a compiled 'bzod' binary to the current directory.${NC}"
+# 2. Install Binary (safe atomic download)
+echo -e "\n${BLUE}[2/8] Installing BZOD binary...${NC}"
+
+ARCH="$(uname -m)"
+case $ARCH in
+    x86_64)  BINARY_NAME="bzod-x86_64-unknown-linux-gnu" ;;
+    aarch64|arm64) BINARY_NAME="bzod-aarch64-unknown-linux-gnu" ;;
+    armv7l)  BINARY_NAME="bzod-armv7-unknown-linux-gnueabihf" ;;
+    *) echo -e "${RED}Unsupported architecture: $ARCH${NC}"; exit 1 ;;
+esac
+
+REPO="thakares/nx9-url-shortener"
+RELEASE_URL="https://github.com/${REPO}/releases/download/v${BZOD_VERSION}/${BINARY_NAME}"
+
+TMP_BINARY=$(mktemp)
+
+echo "Trying GitHub Releases..."
+if curl --retry 5 --retry-delay 2 --retry-connrefused \
+    -L -f -o "${TMP_BINARY}" "${RELEASE_URL}" 2>/dev/null; then
+    echo -e "${GREEN}✓ Downloaded from GitHub Releases${NC}"
+else
+    echo -e "${BLUE}GitHub Releases not available. Trying GHCR...${NC}"
+    if command -v docker >/dev/null 2>&1; then
+        TMP_GHCR=$(mktemp)
+        docker pull ghcr.io/${REPO}:latest >/dev/null 2>&1 || true
+        if docker run --rm --entrypoint cat ghcr.io/${REPO}:latest /usr/local/bin/bzod > "${TMP_GHCR}" 2>/dev/null && [ -s "${TMP_GHCR}" ]; then
+            mv "${TMP_GHCR}" "${TMP_BINARY}"
+            echo -e "${GREEN}✓ Extracted from GHCR${NC}"
+        fi
+    fi
+
+    if [ ! -s "${TMP_BINARY}" ]; then
+        echo -e "${BLUE}Falling back to local build...${NC}"
+        if ! command -v cargo >/dev/null 2>&1; then
+            echo -e "${RED}Neither pre-built binary nor cargo available.${NC}"
+            exit 1
+        fi
+        apt-get install -y pkg-config build-essential
+        cargo build --release
+        cp target/release/bzod "${TMP_BINARY}"
+        echo -e "${GREEN}✓ Built from source${NC}"
+    fi
+fi
+
+# Atomic replace with backup
+if [ -f "${INSTALL_PATH}" ]; then
+    cp "${INSTALL_PATH}" "${INSTALL_PATH}.bak" 2>/dev/null || true
+fi
+
+install -m 755 "${TMP_BINARY}" "${INSTALL_PATH}"
+
+# Verify
+if [ ! -x "${INSTALL_PATH}" ]; then
+    echo -e "${RED}Binary installation failed${NC}"
    exit 1
 fi

-cargo build --release
-echo -e "${GREEN}Release build completed.${NC}"
+"${INSTALL_PATH}" --version >/dev/null && echo -e "${GREEN}✓ Binary verified (--version)${NC}" || {
+    echo -e "${RED}Binary verification failed${NC}"
+    exit 1
+}

-# 4. Install Binary
-echo -e "\n${BLUE}[3/8] Installing binary to ${INSTALL_PATH}...${NC}"
-cp target/release/bzod "${INSTALL_PATH}"
-chmod 755 "${INSTALL_PATH}"
-chown root:root "${INSTALL_PATH}"
-echo -e "${GREEN}Binary installed successfully.${NC}"
+# Verify -V also works
+"${INSTALL_PATH}" -V >/dev/null && echo -e "${GREEN}✓ Binary verified (-V)${NC}" || {
+    echo -e "${RED}Binary -V verification failed${NC}"
+    exit 1
+}

-# 5. Create Dedicated locked-down System User
-echo -e "\n${BLUE}[4/8] Creating dedicated system user '${SERVICE_USER}'...${NC}"
+# Show installed version and verify it matches requested version
+VERSION=$("${INSTALL_PATH}" --version 2>/dev/null | head -n1 || echo "unknown")
+EXPECTED_VERSION="bzod ${BZOD_VERSION}"
+if [ "${VERSION}" != "${EXPECTED_VERSION}" ]; then
+    echo -e "${RED}Version mismatch: expected '${EXPECTED_VERSION}', got '${VERSION}'${NC}"
+    exit 1
+fi
+echo -e "${GREEN}✓ Installed ${VERSION} (${ARCH})${NC}"
+
+# 3. Create System User
+echo -e "\n${BLUE}[3/8] Creating system user '${SERVICE_USER}'...${NC}"
 if ! id -u "${SERVICE_USER}" &>/dev/null; then
    useradd -r -s /usr/sbin/nologin -m -d /var/lib/bzod "${SERVICE_USER}"
-    echo -e "${GREEN}System user '${SERVICE_USER}' created.${NC}"
-else
-    echo "User '${SERVICE_USER}' already exists."
 fi

-# 6. Configure Directory Trees and Permissions
-echo -e "\n${BLUE}[5/8] Setting up configuration and data directories...${NC}"
-mkdir -p "${CONFIG_DIR}"
-mkdir -p "${DATA_DIR}"
+# 4. Setup Directories
+echo -e "\n${BLUE}[4/8] Setting up directories...${NC}"
+mkdir -p "${CONFIG_DIR}" "${DATA_DIR}"
+chown -R "${SERVICE_USER}:${SERVICE_USER}" "/var/lib/bzod"
+chmod 700 "${CONFIG_DIR}"

-# Copy .env file if it exists, otherwise prompt/generate
-if [ -f .env ] && [ ! -f "${ENV_FILE}" ]; then
-    echo "Copying local .env file to ${ENV_FILE}..."
-    cp .env "${ENV_FILE}"
-elif [ ! -f "${ENV_FILE}" ]; then
-    echo "Generating default configuration file at ${ENV_FILE}..."
+# 5. Configuration (preserve on upgrades)
+echo -e "\n${BLUE}[5/8] Configuration...${NC}"
+if [ ! -f "${ENV_FILE}" ]; then
+    echo -e "${BLUE}Generating new secure configuration...${NC}"
    cat <<EOF > "${ENV_FILE}"
 HOST=0.0.0.0
-PORT=8080
+PORT=8654
 DATA_DIR=${DATA_DIR}
 COOKIE_SECURE=true
+RUST_LOG=info
 SESSION_SECRET=$(openssl rand -hex 32)
-ADMIN_USERNAME=admin
-# SHA-256 for bootstrap (Default: admin)
-ADMIN_PASSWORD_SHA256=8c6976e5b5410415bde908bd4dee15dfb167a9c873fc4bb8a81f6f2ab448a918
-LINK_CHECK_INTERVAL_MINS=60
-AGGREGATION_INTERVAL_MINS=60
-DATA_RETENTION_DAYS=365
 EOF
+    chmod 600 "${ENV_FILE}"
+    chown root:"${SERVICE_USER}" "${ENV_FILE}"
+else
+    echo -e "${GREEN}Existing configuration preserved${NC}"
 fi

-chmod 600 "${ENV_FILE}"
-chown -R root:"${SERVICE_USER}" "${CONFIG_DIR}"
-chown -R "${SERVICE_USER}":"${SERVICE_USER}" /var/lib/bzod
-echo -e "${GREEN}Directories and permission parameters configured.${NC}"
-
-# 7. Initialise DB as the service user (avoids file permission conflicts)
-echo -e "\n${BLUE}[6/8] Initialising databases...${NC}"
-sudo -u "${SERVICE_USER}" "${INSTALL_PATH}" init-db --data-dir "${DATA_DIR}"
-echo -e "${GREEN}Databases initialised.${NC}"
-
-# 8. Set Up Systemd Service
-echo -e "\n${BLUE}[7/8] Installing systemd service unit...${NC}"
+# 6. Systemd Service
+echo -e "\n${BLUE}[6/8] Installing hardened systemd service...${NC}"
 cat <<EOF > "${SYSTEMD_UNIT}"
 [Unit]
-Description=BZOD - Personal URL Shortener & Landing Page Platform
-After=network.target
+Description=BZOD - Privacy-First URL Shortener & Landing Page Platform
+After=network-online.target
+Wants=network-online.target

 [Service]
 Type=simple
@@ -107,11 +159,12 @@ User=${SERVICE_USER}
 Group=${SERVICE_USER}
 WorkingDirectory=/var/lib/bzod
 EnvironmentFile=${ENV_FILE}
-ExecStart=${INSTALL_PATH} serve --host 0.0.0.0 --port 8080 --data-dir ${DATA_DIR}
+ExecStart=${INSTALL_PATH} serve
+
 Restart=on-failure
 RestartSec=5s

-# Hardening / Sandboxing options for security
+# Security Hardening
 ProtectSystem=strict
 ProtectHome=yes
 PrivateTmp=yes
@@ -119,6 +172,10 @@ PrivateDevices=yes
 ProtectKernelTunables=yes
 ProtectKernelModules=yes
 ProtectControlGroups=yes
+ProtectHostname=yes
+RestrictSUIDSGID=yes
+LockPersonality=yes
+NoNewPrivileges=yes
 ReadWritePaths=/var/lib/bzod

 [Install]
@@ -127,21 +184,68 @@ EOF

 chmod 644 "${SYSTEMD_UNIT}"
 systemctl daemon-reload
-echo -e "${GREEN}Systemd service registered.${NC}"

-# 9. Enable and Start the Service
-echo -e "\n${BLUE}[8/8] Starting BZOD service...${NC}"
-systemctl enable bzod
-systemctl restart bzod
+# 7. Initialize & Start
+echo -e "\n${BLUE}[7/8] Initializing and starting service...${NC}"

-sleep 2
-if systemctl is-active --quiet bzod; then
-    echo -e "${GREEN}BZOD service is running successfully!${NC}"
-    echo -e "\n${BLUE}=== Deployment Completed Successfully ===${NC}"
-    echo -e "You can access BZOD at http://localhost:8080"
-    echo -e "Admin Login Dashboard is at http://localhost:8080/admin"
-    echo -e "System service logs: journalctl -u bzod -f"
-    echo -e "To change the default admin password, run: bzod create-admin --data-dir ${DATA_DIR}"
+# Database creation and migration is handled automatically by 'bzod serve'
+if [ -f "${DATA_DIR}/admin/admin.db" ] || [ -f "${DATA_DIR}/admin.db" ]; then
+    echo -e "${GREEN}✓ Existing database detected (upgrade mode)${NC}"
+
+    # Pre-upgrade: stop service and backup databases
+    if systemctl is-active --quiet bzod 2>/dev/null; then
+        echo -e "${BLUE}  Stopping BZOD for safe database backup...${NC}"
+        systemctl stop bzod
+    fi
+
+    BACKUP_DIR="/var/lib/bzod/pre-upgrade-backup-v${BZOD_VERSION}"
+    mkdir -p "${BACKUP_DIR}"
+    cp -a "${DATA_DIR}" "${BACKUP_DIR}/data" 2>/dev/null || true
+    cp "${ENV_FILE}" "${BACKUP_DIR}/bzod.env" 2>/dev/null || true
+    echo -e "${GREEN}  ✓ Pre-upgrade backup created at ${BACKUP_DIR}${NC}"
 else
-    echo -e "${RED}Error: BZOD service failed to start. Check logs using: journalctl -u bzod -n 50${NC}"
+    echo -e "${GREEN}✓ Fresh installation (databases will be created on first start)${NC}"
 fi
+
+systemctl enable --now bzod
+
+# 8. Validation + Rollback
+sleep 3
+
+if ! systemctl is-active --quiet bzod; then
+    echo -e "${RED}Service failed to start! Rolling back...${NC}"
+    if [ -f "${INSTALL_PATH}.bak" ]; then
+        install -m 755 "${INSTALL_PATH}.bak" "${INSTALL_PATH}"
+        systemctl restart bzod || true
+    fi
+    journalctl -u bzod -n 50 --no-pager
+    exit 1
+fi
+
+# Clean up backup on success
+rm -f "${INSTALL_PATH}.bak" 2>/dev/null || true
+
+# Soft health check
+if command -v curl >/dev/null 2>&1; then
+    if curl -fsS http://127.0.0.1:8654/status >/dev/null 2>&1; then
+        echo -e "${GREEN}✓ HTTP health check passed${NC}"
+    else
+        echo -e "${BLUE}✓ Service is running (systemd healthy)${NC}"
+    fi
+fi
+
+# Final Message
+IP=$(hostname -I | awk '{print $1}' | head -n1)
+echo -e "\n${GREEN}=== BZOD Deployed Successfully! ===${NC}"
+echo -e "🌐 Web UI:     http://${IP}:8654"
+echo -e "🔑 Admin:      http://${IP}:8654/admin"
+echo -e "🖥 Architecture: ${ARCH}"
+echo -e "📦 Version: ${VERSION}"
+echo -e "\nNext step (first install):"
+echo -e "   sudo -u bzod bzod create-admin"
+echo -e "\nCommands:"
+echo -e "   journalctl -u bzod -f"
+echo -e "   bzod doctor"
+echo -e "   systemctl status bzod"
+
+echo -e "\n${GREEN}Enjoy your lightweight, privacy-first, self-hosted URL shortener!${NC}"
@@ -1,67 +1,63 @@
 name: app-bzod
-
 services:
  bzod:
-    image: nx9-url-shortener:v0.1.0
-
    build:
-      context: .
+      context: /DATA/AppData/bzod
      dockerfile: Dockerfile
-
+    cpu_shares: 90
+    command: []
    container_name: bzod
-
+    deploy:
+      resources:
+        limits:
+          memory: 31940M
    environment:
      - COOKIE_SECURE=false
      - DATA_DIR=/app/data
      - HOST=0.0.0.0
      - PORT=8654
      - RUST_LOG=info
-
+    hostname: bzod
+    image: nx9-url-shortener:v0.6.0
    ports:
      - mode: ingress
        target: 8654
        published: "8654"
        protocol: tcp
-
    restart: unless-stopped
-
    volumes:
      - type: bind
        source: /DATA/AppData/bzod/data
        target: /app/data
        bind:
          create_host_path: true
-
      - type: bind
        source: /DATA/AppData/bzod/config
        target: /app/config
        bind:
          create_host_path: true
-
+      - type: bind
+        source: /DATA/AppData/bzod/www
+        target: /app/www
+      - type: bind
+        source: /DATA/AppData/bzod/images
+        target: /app/images
+    devices: []
+    cap_add: []
    networks:
      - default
-
-    hostname: bzod
    privileged: false
-
-    cpu_shares: 90
-
-    deploy:
-      resources:
-        limits:
-          memory: 31940M
-
 networks:
  default:
    name: app_default
-
 x-casaos:
-  hostname: ""
-  scheme: http
-  index: /
-  port_map: "8654"
  author: self
  category: self
+  hostname: ""
  icon: ""
+  index: /
+  is_uncontrolled: false
+  port_map: "8654"
+  scheme: http
  title:
-    custom: nx9-url-shortener
+    custom: nx9-url-shortener
@@ -0,0 +1,888 @@
+# BZOD Administrator Guide
+
+Version: v0.6.0
+
+---
+
+# Introduction
+
+This guide is intended for BZOD administrators responsible for operating, maintaining, and managing a BZOD instance.
+
+It covers:
+
+* Administrator authentication
+* User management
+* Quotas
+* Sessions
+* Moderation
+* Slug ownership
+* Analytics
+* Audit logs
+* Backup and recovery
+* Health monitoring
+* Operational best practices
+
+---
+
+# Administrator Role
+
+Administrators have full platform control.
+
+Administrative capabilities include:
+
+* Create users
+* Modify users
+* Disable users
+* Delete users
+* Reset passwords
+* Manage quotas
+* Review analytics
+* Moderate content
+* Transfer slug ownership
+* Manage backups
+* Review audit logs
+* Monitor system health
+
+Administrators cannot bypass audit logging.
+
+All administrative actions are recorded.
+
+---
+
+# Login
+
+Administrative login is available at:
+
+```text
+/login
+```
+
+Successful login redirects to:
+
+```text
+/admin
+```
+
+Authentication uses:
+
+```text
+users.db
+```
+
+Sessions are stored in:
+
+```text
+users.db.sessions
+```
+
+Cookie name:
+
+```text
+bzod_session
+```
+
+---
+
+# Administrative Dashboard
+
+Route:
+
+```text
+/admin
+```
+
+The dashboard provides a high-level overview of platform activity.
+
+Metrics include:
+
+* Total Users
+* Active Users
+* Total URLs
+* Total Landing Pages
+* Active Sessions
+* API Tokens
+* Storage Usage
+* Moderation Events
+* Recent Audit Events
+
+Quick actions include:
+
+* Create User
+* View Sessions
+* View Audit Logs
+* Create Backup
+* Review Health Status
+
+---
+
+# User Management
+
+## Users List
+
+Route:
+
+```text
+/admin/users
+```
+
+Displays:
+
+* User ID
+* Username
+* Status
+* Account Type
+* Creation Date
+
+Available actions:
+
+* View
+* Edit
+* Disable
+* Enable
+* Reset Password
+* Delete
+
+---
+
+## Create User
+
+Route:
+
+```text
+/admin/users/new
+```
+
+Fields:
+
+* Username
+* Password
+* Account Type
+* Quota Limits
+
+Supported account types:
+
+```text
+admin
+standard
+```
+
+Reserved usernames cannot be used.
+
+Examples:
+
+```text
+admin
+legacy_admin
+system
+root
+administrator
+```
+
+---
+
+## User Detail Page
+
+Route:
+
+```text
+/admin/users/{id}
+```
+
+Displays:
+
+### Profile
+
+* User ID
+* Username
+* Status
+* Account Type
+* Created Date
+
+### Usage Statistics
+
+* URL Count
+* Landing Page Count
+* Visit Count
+* Storage Usage
+* API Token Count
+* Active Sessions
+
+### Quotas
+
+* Maximum URLs
+* Maximum Pages
+* Maximum Storage
+* Maximum Tokens
+
+### Sessions
+
+List of active sessions.
+
+### API Tokens
+
+List of active tokens.
+
+---
+
+## Edit User
+
+Route:
+
+```text
+/admin/users/{id}/edit
+```
+
+Administrators may:
+
+* Change status
+* Change account type
+* Modify quotas
+
+---
+
+## Reset Password
+
+Route:
+
+```text
+/admin/users/{id}/password
+```
+
+Creates a new password hash and invalidates existing sessions.
+
+Audit event generated:
+
+```text
+password_reset
+```
+
+---
+
+## Disable User
+
+Route:
+
+```text
+/admin/users/{id}/disable
+```
+
+Effects:
+
+* User login disabled
+* Existing sessions revoked
+* API access denied
+
+Audit event generated:
+
+```text
+user_disabled
+```
+
+---
+
+## Enable User
+
+Route:
+
+```text
+/admin/users/{id}/enable
+```
+
+Restores account access.
+
+Audit event generated:
+
+```text
+user_enabled
+```
+
+---
+
+## Delete User
+
+Route:
+
+```text
+/admin/users/{id}/delete
+```
+
+Deletion performs:
+
+1. Session revocation
+2. API token removal
+3. Content removal
+4. Analytics removal
+5. Slug release
+6. User database deletion
+
+Audit event generated:
+
+```text
+user_deleted
+```
+
+---
+
+# Session Management
+
+Route:
+
+```text
+/admin/sessions
+```
+
+Displays all active platform sessions.
+
+Information displayed:
+
+* User ID
+* Username
+* Session Identifier
+* Created Time
+* Expiry Time
+* IP Address
+* User Agent
+
+---
+
+## Revoke Session
+
+Individual sessions can be revoked.
+
+Effects:
+
+* Session removed immediately
+* User forced to reauthenticate
+
+---
+
+## Revoke All Sessions
+
+Administrators may invalidate all active sessions.
+
+Useful after:
+
+* Password compromise
+* Security incidents
+* Large configuration changes
+
+---
+
+# Quota Management
+
+Route:
+
+```text
+/admin/quotas
+```
+
+Quotas limit user resource consumption.
+
+Available limits:
+
+```text
+max_urls
+max_pages
+max_storage_mb
+max_api_tokens
+```
+
+---
+
+## Quota Reconciliation
+
+Administrators can execute:
+
+```text
+quota_reconcile
+```
+
+Purpose:
+
+* Detect counter drift
+* Recount resources
+* Repair quota usage
+
+Common causes:
+
+* Manual database modifications
+* Failed migrations
+* Interrupted operations
+
+---
+
+# Moderation
+
+Route:
+
+```text
+/admin/moderation
+```
+
+Moderation allows administrators to manage abuse and policy violations.
+
+---
+
+## Flag Content
+
+Marks content for review.
+
+Audit event:
+
+```text
+content_flagged
+```
+
+---
+
+## Disable Content
+
+Disabled content returns:
+
+```http
+410 Gone
+```
+
+Affected endpoints:
+
+```text
+/{slug}
+/p/{slug}
+/api/qr/{slug}.png
+/api/qr/{slug}.svg
+```
+
+Audit event:
+
+```text
+content_disabled
+```
+
+---
+
+## Enable Content
+
+Restores functionality.
+
+Audit event:
+
+```text
+content_enabled
+```
+
+---
+
+## Delete Content
+
+Permanently removes content.
+
+Audit event:
+
+```text
+content_deleted
+```
+
+---
+
+# Slug Management
+
+Route:
+
+```text
+/admin/slugs
+```
+
+Displays platform-wide slug ownership.
+
+Information includes:
+
+* Slug
+* Owner
+* Type
+* Status
+* Creation Date
+
+---
+
+## Slug Types
+
+Supported types:
+
+```text
+url
+page
+```
+
+---
+
+## Transfer Ownership
+
+Administrators may transfer ownership.
+
+Workflow:
+
+1. Validate recipient quota.
+2. Copy content.
+3. Update ownership.
+4. Update global slug registry.
+5. Write audit record.
+
+Audit event:
+
+```text
+slug_transfer
+```
+
+Analytics are preserved.
+
+---
+
+# Analytics
+
+Administrators can access analytics for any managed resource.
+
+---
+
+## URL Analytics
+
+Route:
+
+```text
+/admin/analytics/url/{id}
+```
+
+Displays:
+
+* Total Visits
+* Unique Visitors
+* Referrers
+* Browsers
+* Countries
+* Visit Timeline
+
+---
+
+## Page Analytics
+
+Route:
+
+```text
+/admin/analytics/page/{id}
+```
+
+Displays identical metrics for landing pages.
+
+---
+
+## User Analytics
+
+Administrators can review user-level analytics.
+
+Route:
+
+```text
+/analytics
+```
+
+Includes:
+
+* Top Links
+* Top Pages
+* Referrers
+* Browsers
+* Countries
+* Recent Visits
+
+---
+
+# Audit Logs
+
+Route:
+
+```text
+/admin/audit
+```
+
+All administrative actions are recorded.
+
+Searchable event types include:
+
+```text
+login
+logout
+failed_login
+user_created
+user_deleted
+user_disabled
+user_enabled
+password_reset
+quota_updated
+slug_transfer
+content_flagged
+content_disabled
+backup_created
+restore_executed
+```
+
+Audit logs should be reviewed regularly.
+
+---
+
+# Backup Management
+
+Route:
+
+```text
+/admin/backups
+```
+
+Provides web-based backup operations.
+
+---
+
+## Create Backup
+
+Creates a platform snapshot.
+
+Includes:
+
+```text
+users.db
+system.db
+tenant databases
+```
+
+Audit event:
+
+```text
+backup_created
+```
+
+---
+
+## Download Backup
+
+Allows local storage of backup archives.
+
+Recommended frequency:
+
+```text
+Daily
+```
+
+---
+
+## Restore Backup
+
+Restores a selected backup archive.
+
+Audit event:
+
+```text
+restore_executed
+```
+
+Always test restores before production use.
+
+---
+
+## Delete Backup
+
+Removes backup archives from storage.
+
+---
+
+# Health Dashboard
+
+Route:
+
+```text
+/admin/health
+```
+
+Provides operational diagnostics.
+
+Displays:
+
+* Database Status
+* WAL Status
+* Storage Utilization
+* Backup Status
+* Health Check Results
+* Quota Reconciliation Results
+
+---
+
+## Database Health
+
+Checks:
+
+```text
+users.db
+system.db
+content.db
+analytics.db
+```
+
+Reports:
+
+```text
+healthy
+warning
+error
+```
+
+---
+
+## Storage Monitoring
+
+Shows:
+
+* Total Storage
+* Free Storage
+* Database Sizes
+* Backup Sizes
+
+---
+
+# Security Administration
+
+## Password Policies
+
+Recommendations:
+
+* Minimum 12 characters
+* Unique passwords
+* Password manager usage
+
+---
+
+## Session Management
+
+Recommended actions:
+
+* Revoke old sessions
+* Review active sessions
+* Remove inactive users
+
+---
+
+## CSRF Protection
+
+All administrative forms require valid CSRF tokens.
+
+Invalid requests return:
+
+```http
+403 Forbidden
+```
+
+---
+
+## Audit Reviews
+
+Recommended review schedule:
+
+| Event Type        | Frequency |
+| ----------------- | --------- |
+| Failed Logins     | Daily     |
+| User Creation     | Weekly    |
+| Slug Transfers    | Weekly    |
+| Backup Events     | Daily     |
+| Moderation Events | Weekly    |
+
+---
+
+# Disaster Recovery
+
+Recommended workflow:
+
+1. Stop BZOD.
+2. Create backup copy.
+3. Restore archive.
+4. Verify databases.
+5. Run integrity checks.
+6. Restart service.
+
+---
+
+# Operational Best Practices
+
+Recommended:
+
+* Enable HTTPS
+* Run daily backups
+* Monitor disk usage
+* Review audit logs
+* Keep binaries updated
+* Test restore procedures regularly
+
+Avoid:
+
+* Manual database modifications
+* Direct deletion of tenant databases
+* Disabling audit logging
+
+---
+
+# Troubleshooting
+
+## User Cannot Login
+
+Check:
+
+* User status
+* Session validity
+* Password reset history
+
+---
+
+## Slug Already Exists
+
+Check:
+
+```text
+/admin/slugs
+```
+
+for ownership conflicts.
+
+---
+
+## Analytics Missing
+
+Verify:
+
+* Analytics worker running
+* Analytics database present
+* Event queue processing
+
+---
+
+## Backup Failure
+
+Check:
+
+* Free disk space
+* File permissions
+* Backup destination path
+
+---
+
+# Summary
+
+The BZOD administration system provides:
+
+* Centralized user management
+* Quotas and session controls
+* Moderation and slug ownership management
+* Analytics visibility
+* Audit logging
+* Backup and restore capabilities
+* Health monitoring
+
+while maintaining strong tenant isolation and a SQLite-native operational model.
+
+---
+
+End of Document.
@@ -0,0 +1,391 @@
+# BZOD REST API
+
+> Programmatic access to URLs, Landing Pages, QR Codes, Analytics, and Audit Logs.
+
+## Overview
+
+The BZOD REST API allows automation and integration with external systems such as:
+
+* Home Assistant
+* Shell Scripts
+* CI/CD Pipelines
+* Monitoring Systems
+* Internal Applications
+* Self-hosted Services
+
+All API endpoints require authentication using an API Token generated from:
+
+```text
+Admin Dashboard → Settings → REST API Tokens
+```
+
+---
+
+# Authentication
+
+Generate an API token from the Admin Dashboard.
+
+Example token:
+
+```text
+bzo_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+```
+
+Pass the token using the `Authorization` header.
+
+## Example
+
+```bash
+curl \
+  -H "Authorization: bzo_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" \
+  https://your-domain.com/api/v1/stats
+```
+
+---
+
+# Base URL
+
+```text
+https://your-domain.com/api/v1
+```
+
+Example:
+
+```text
+https://bzo.in/api/v1
+```
+
+---
+
+# Response Format
+
+Successful responses:
+
+```json
+{
+  "success": true,
+  "data": {}
+}
+```
+
+Error responses:
+
+```json
+{
+  "success": false,
+  "error": "Invalid API token"
+}
+```
+
+---
+
+# URL Management
+
+## List URLs
+
+```http
+GET /api/v1/urls
+```
+
+### Example
+
+```bash
+curl \
+  -H "Authorization: TOKEN" \
+  https://your-domain.com/api/v1/urls
+```
+
+---
+
+## Create URL
+
+```http
+POST /api/v1/urls
+```
+
+### Request
+
+```json
+{
+  "code": "rust",
+  "target_url": "https://www.rust-lang.org",
+  "description": "Rust Language"
+}
+```
+
+### Example
+
+```bash
+curl \
+  -X POST \
+  -H "Authorization: TOKEN" \
+  -H "Content-Type: application/json" \
+  -d '{
+        "code":"rust",
+        "target_url":"https://www.rust-lang.org"
+      }' \
+  https://your-domain.com/api/v1/urls
+```
+
+---
+
+## Get URL
+
+```http
+GET /api/v1/urls/{uuid}
+```
+
+Example:
+
+```http
+GET /api/v1/urls/5d4d9e98-7cb7-4c97-9a0a-123456789abc
+```
+
+---
+
+## Update URL
+
+```http
+PUT /api/v1/urls/{uuid}
+```
+
+---
+
+## Delete URL
+
+```http
+DELETE /api/v1/urls/{uuid}
+```
+
+---
+
+## URL Preview
+
+```http
+GET /api/v1/urls/{uuid}/preview
+```
+
+Returns rendered metadata used by preview cards.
+
+---
+
+# Landing Pages
+
+## List Pages
+
+```http
+GET /api/v1/pages
+```
+
+---
+
+## Create Page
+
+```http
+POST /api/v1/pages
+```
+
+### Example Request
+
+```json
+{
+  "title": "My Product",
+  "slug": "product",
+  "description": "Product Landing Page",
+  "content": "<h1>Hello World</h1>"
+}
+```
+
+---
+
+## Get Page
+
+```http
+GET /api/v1/pages/{uuid}
+```
+
+---
+
+## Update Page
+
+```http
+PUT /api/v1/pages/{uuid}
+```
+
+---
+
+## Delete Page
+
+```http
+DELETE /api/v1/pages/{uuid}
+```
+
+---
+
+# Analytics
+
+## Global Statistics
+
+```http
+GET /api/v1/stats
+```
+
+Returns overall platform metrics.
+
+Example response:
+
+```json
+{
+  "total_urls": 125,
+  "total_pages": 12,
+  "total_clicks": 8431,
+  "total_qr_scans": 241
+}
+```
+
+---
+
+## URL Statistics
+
+```http
+GET /api/v1/stats/url/{uuid}
+```
+
+Returns analytics for a single URL.
+
+---
+
+## Landing Page Statistics
+
+```http
+GET /api/v1/stats/page/{uuid}
+```
+
+Returns analytics for a single landing page.
+
+---
+
+# QR Codes
+
+## Download QR Code
+
+```http
+GET /api/v1/qr/{code}
+```
+
+Example:
+
+```http
+GET /api/v1/qr/rust
+```
+
+Returns QR image.
+
+---
+
+# Bulk Operations
+
+## Bulk QR Export
+
+```http
+POST /api/v1/bulk/qr
+```
+
+Generate QR codes for multiple URLs.
+
+---
+
+## Bulk URL Operations
+
+```http
+POST /api/v1/bulk/url
+```
+
+Bulk create, update, or manage URLs.
+
+---
+
+# Audit Log
+
+## List Audit Events
+
+```http
+GET /api/v1/audit
+```
+
+Returns administrative activity history.
+
+Example response:
+
+```json
+[
+  {
+    "event": "url_created",
+    "user": "admin",
+    "timestamp": "2026-06-17T14:30:00Z"
+  }
+]
+```
+
+---
+
+# HTTP Status Codes
+
+| Code | Description           |
+| ---- | --------------------- |
+| 200  | Success               |
+| 201  | Created               |
+| 400  | Invalid Request       |
+| 401  | Authentication Failed |
+| 403  | Access Denied         |
+| 404  | Resource Not Found    |
+| 409  | Conflict              |
+| 500  | Internal Server Error |
+
+---
+
+# Security Notes
+
+* API tokens are displayed only once during creation.
+* Tokens are stored as hashes and cannot be recovered.
+* Revoke unused tokens immediately.
+* Always use HTTPS.
+* Never embed API tokens in public repositories.
+
+---
+
+# Example: Create URL From Shell Script
+
+```bash
+TOKEN="bzo_xxxxxxxxxxxxxxxxx"
+
+curl \
+  -X POST \
+  -H "Authorization: ${TOKEN}" \
+  -H "Content-Type: application/json" \
+  -d '{
+        "code":"example",
+        "target_url":"https://example.com"
+      }' \
+  https://your-domain.com/api/v1/urls
+```
+
+---
+
+# API Stability
+
+The BZOD API follows semantic versioning.
+
+Current API namespace:
+
+```text
+/api/v1
+```
+
+Future breaking changes will be introduced under a new versioned namespace.
+
+Example:
+
+```text
+/api/v2
+```
@@ -0,0 +1,667 @@
+# BZOD Architecture Guide
+
+Version: v0.6.0
+
+---
+
+# Overview
+
+BZOD is a self-hosted multi-user URL management platform written in Rust.
+
+The platform combines:
+
+* URL shortening
+* Landing pages
+* QR code generation
+* Analytics
+* User management
+* Moderation
+* Audit logging
+* Backup & restore
+* Disaster recovery
+
+into a single deployable binary powered entirely by SQLite.
+
+BZOD is designed around operational simplicity, tenant isolation, and long-term maintainability.
+
+---
+
+# Architectural Goals
+
+The primary design goals are:
+
+1. Self-hosted first
+2. SQLite-first architecture
+3. Multi-user operation
+4. Tenant isolation
+5. Simple deployment
+6. Minimal dependencies
+7. Easy backup and recovery
+8. No vendor lock-in
+
+---
+
+# High-Level Architecture
+
+```text
+                     ┌─────────────┐
+                     │   Browser   │
+                     └──────┬──────┘
+                            │
+                            ▼
+                 ┌────────────────────┐
+                 │     Axum Router    │
+                 └─────────┬──────────┘
+                           │
+      ┌────────────────────┼────────────────────┐
+      │                    │                    │
+      ▼                    ▼                    ▼
+
+  users.db           system.db         User Databases
+
+  Users              Global Slugs      content.db
+  Sessions           Audit Events      analytics.db
+  Quotas             Moderation
+  API Tokens         Settings
+```
+
+---
+
+# Runtime Components
+
+## Web Layer
+
+Location:
+
+```text
+src/web/
+```
+
+Responsible for:
+
+* HTTP routing
+* Dashboard rendering
+* Form handling
+* Authentication checks
+* Redirect handling
+* REST API endpoints
+
+Major modules:
+
+```text
+admin/          (modular feature directory)
+  auth.rs       (authentication and session handling)
+  dashboard.rs  (dashboard rendering)
+  urls.rs       (URL management handlers)
+  pages.rs      (landing page management handlers)
+  analytics.rs  (analytics and export handlers)
+  settings.rs   (settings and configuration handlers)
+  users.rs      (user management handlers)
+  sessions.rs   (session administration)
+  quotas.rs     (quota management)
+  health.rs     (health diagnostics)
+  backups.rs    (backup and restore handlers)
+  api_keys.rs   (API key management)
+  audit.rs      (audit log handlers)
+  moderation.rs (content moderation handlers)
+  mod.rs        (module exports and shared helpers)
+api.rs
+pages.rs
+redirect.rs
+qr.rs
+system.rs
+multi_user.rs
+routes.rs
+```
+
+---
+
+## Authentication Layer
+
+Location:
+
+```text
+src/auth/
+```
+
+Responsible for:
+
+* Password hashing
+* Session validation
+* Cookie management
+* CSRF protection
+* Authorization
+
+Modules:
+
+```text
+csrf.rs
+middleware.rs
+password.rs
+session.rs
+```
+
+Authentication technologies:
+
+* Argon2id password hashing
+* Session cookies
+* CSRF tokens
+* RBAC checks
+
+---
+
+## Database Layer
+
+Location:
+
+```text
+src/db/
+```
+
+Responsible for:
+
+* Schema creation
+* Migrations
+* Database access
+* Analytics storage
+* User management
+
+Modules:
+
+```text
+admin.rs
+analytics.rs
+audit_events.rs
+content.rs
+migrations.rs
+sqlite.rs
+users.rs
+```
+
+---
+
+# Database Architecture
+
+BZOD uses multiple SQLite databases rather than a single monolithic database.
+
+This approach provides:
+
+* Better isolation
+* Easier backup
+* Simpler disaster recovery
+* Reduced risk of cross-user data leakage
+
+---
+
+## users.db
+
+Purpose:
+
+Central identity and account database.
+
+Contains:
+
+```text
+users
+sessions
+api_tokens
+quotas
+```
+
+Stores:
+
+* User accounts
+* Password hashes
+* Session records
+* API tokens
+* Quota information
+
+---
+
+## system.db
+
+Purpose:
+
+Global platform metadata.
+
+Contains:
+
+```text
+global_slugs
+audit_events
+moderation_events
+reserved_slugs
+settings
+slug_history
+```
+
+Stores:
+
+* Global slug ownership
+* Audit records
+* Moderation actions
+* Platform settings
+* Slug transfers
+
+---
+
+## Tenant Databases
+
+Each user receives isolated databases.
+
+Directory structure:
+
+```text
+users/
+└── <user_id>/
+    ├── content.db
+    └── analytics.db
+```
+
+---
+
+### content.db
+
+Stores:
+
+* URLs
+* Landing pages
+* Metadata
+
+---
+
+### analytics.db
+
+Stores:
+
+* Visits
+* Referrers
+* QR scans
+* Browser information
+* Analytics aggregates
+
+---
+
+# Multi-User Architecture
+
+BZOD v0.5.0 introduced complete tenant isolation.
+
+Each user owns:
+
+```text
+content.db
+analytics.db
+```
+
+Users cannot directly access:
+
+* Other users' URLs
+* Other users' landing pages
+* Other users' analytics
+
+The administrator accesses all tenants through controlled administrative interfaces.
+
+---
+
+# Global Slug Namespace
+
+All public URLs are tracked in:
+
+```text
+system.db -> global_slugs
+```
+
+Purpose:
+
+Prevent collisions across users.
+
+Example:
+
+```text
+User A owns:
+
+https://bzo.in/!office
+
+User B cannot create:
+
+https://bzo.in/!office
+```
+
+This guarantees global uniqueness.
+
+---
+
+# Request Lifecycle
+
+## URL Redirect
+
+Request:
+
+```text
+GET /abc123
+```
+
+Flow:
+
+```text
+Browser
+  ↓
+Axum Router
+  ↓
+global_slugs lookup
+  ↓
+Locate owner database
+  ↓
+Resolve URL
+  ↓
+Validate destination
+  ↓
+Record analytics
+  ↓
+301 Redirect (with safe Location header construction)
+```
+
+---
+
+## Landing Page
+
+Request:
+
+```text
+GET /p/demo
+```
+
+Flow:
+
+```text
+Browser
+  ↓
+Router
+  ↓
+global_slugs lookup
+  ↓
+Tenant content.db lookup
+  ↓
+Render page
+```
+
+---
+
+## QR Generation
+
+Request:
+
+```text
+GET /api/qr/demo.svg
+```
+
+Flow:
+
+```text
+Router
+  ↓
+global_slugs lookup
+  ↓
+Generate QR
+  ↓
+Return SVG
+```
+
+---
+
+# Analytics Pipeline
+
+Location:
+
+```text
+src/analytics/
+```
+
+Components:
+
+```text
+events.rs
+queue.rs
+worker.rs
+aggregate.rs
+location.rs
+```
+
+Responsibilities:
+
+* Visit tracking
+* QR tracking
+* Browser detection
+* Referrer parsing
+* Aggregation
+
+---
+
+# Background Jobs
+
+Location:
+
+```text
+src/jobs/
+```
+
+Jobs:
+
+## aggregate.rs
+
+Analytics aggregation.
+
+## backup.rs
+
+Automated backups.
+
+## expiry.rs
+
+Expired content cleanup.
+
+## retention.rs
+
+Retention policy enforcement.
+
+## healthcheck.rs
+
+System health validation.
+
+## quota_reconcile.rs
+
+Quota consistency verification.
+
+---
+
+# Services Layer
+
+Location:
+
+```text
+src/services/
+```
+
+Purpose:
+
+Business logic abstraction.
+
+Modules:
+
+```text
+api_keys.rs
+audit.rs
+bulk.rs
+landing_pages.rs
+qr.rs
+shortener.rs
+```
+
+This layer separates business rules from HTTP handlers.
+
+---
+
+# CLI Architecture
+
+Location:
+
+```text
+src/cli/
+```
+
+The CLI and Web UI share the same internal services.
+
+Examples:
+
+```bash
+bzod create-admin
+bzod create-user
+bzod backup
+bzod restore
+bzod doctor
+bzod migrate
+```
+
+This avoids duplicate logic between administration methods.
+
+---
+
+# Security Model
+
+Security mechanisms:
+
+## Authentication
+
+* Argon2id password hashes
+* Session cookies
+
+## Authorization
+
+* RBAC
+* Administrative permission checks
+
+## CSRF Protection
+
+* Form tokens
+* Request validation
+
+## Tenant Isolation
+
+* Separate databases
+* Controlled access paths
+
+## Audit Logging
+
+All critical operations are recorded.
+
+Examples:
+
+* Login attempts
+* User creation
+* Password resets
+* Slug transfers
+* Moderation actions
+
+---
+
+# Backup & Recovery
+
+BZOD is designed for SQLite-first recovery.
+
+Backup targets:
+
+```text
+users.db
+system.db
+admin/
+users/*
+```
+
+Capabilities:
+
+* Full backups
+* Restore operations
+* Upgrade migrations
+* Disaster recovery validation
+
+---
+
+# Testing Architecture
+
+Location:
+
+```text
+tests/
+```
+
+Coverage includes:
+
+* Authentication
+* Authorization
+* User management
+* Analytics
+* Backups
+* Disaster recovery
+* Routing
+* Security
+* Concurrency
+* Upgrade validation
+* Multi-user isolation
+
+The project includes comprehensive automated test coverage spanning unit, integration, security, and end-to-end tests.
+
+---
+
+# Deployment Models
+
+Supported deployments:
+
+## Native
+
+```bash
+cargo build --release
+./bzod serve
+```
+
+## Systemd
+
+```text
+bzod.service
+```
+
+## Docker
+
+```text
+Dockerfile
+docker-compose.yml
+```
+
+---
+
+# Future Architecture Direction
+
+Planned for future releases:
+
+* Geo analytics
+* OpenAPI generation
+* SSO integration
+* Multi-organization support
+* Advanced reporting
+* Distributed analytics aggregation
+
+---
+
+# Summary
+
+BZOD is built around a simple principle:
+
+> Keep deployment simple, keep data local, keep users isolated, and keep recovery easy.
+
+The platform achieves this through:
+
+* Rust
+* Axum
+* SQLite
+* Tenant isolation
+* Multi-database architecture
+* Strong automated validation
+* Operational simplicity
@@ -0,0 +1,584 @@
+# Backup & Restore Guide
+
+Version: v0.6.0
+Applies To: BZOD Multi-User Platform
+
+---
+
+# Overview
+
+BZOD provides built-in backup and recovery functionality for both single-user and multi-user deployments.
+
+The backup architecture is designed to support:
+
+* Full platform backups
+* Individual tenant backups
+* Disaster recovery
+* Upgrade safety
+* Migration validation
+* Data integrity verification
+
+All production deployments should maintain regular backups before performing upgrades, maintenance, or administrative operations.
+
+---
+
+# Database Architecture
+
+BZOD stores data across multiple SQLite databases.
+
+## Core Databases
+
+```text
+data/
+├── users.db
+├── system.db
+└── users/
+```
+
+### users.db
+
+Stores:
+
+* User accounts
+* Password hashes
+* Account status
+* Roles
+* Sessions
+* Quotas
+* API tokens
+
+### system.db
+
+Stores:
+
+* Global slug registry
+* Reserved slugs
+* Slug ownership history
+* Audit events
+* Moderation events
+* System settings
+
+---
+
+## Tenant Databases
+
+Each tenant owns isolated content and analytics databases.
+
+```text
+data/users/{user_id}/
+├── content.db
+└── analytics.db
+```
+
+### content.db
+
+Stores:
+
+* Short URLs
+* Landing pages
+* Metadata
+* Tags
+* QR code configuration
+
+### analytics.db
+
+Stores:
+
+* Visit events
+* Referrers
+* Browser information
+* Country information
+* Aggregated statistics
+
+---
+
+# Backup Types
+
+## Full Platform Backup
+
+Creates a complete snapshot of the entire BZOD installation.
+
+Includes:
+
+```text
+users.db
+system.db
+all tenant content.db files
+all tenant analytics.db files
+```
+
+Recommended for:
+
+* Daily scheduled backups
+* Upgrades
+* Server migration
+* Disaster recovery
+
+---
+
+## User Backup
+
+Creates a backup of a single tenant.
+
+Includes:
+
+```text
+content.db
+analytics.db
+```
+
+Recommended for:
+
+* User export
+* User migration
+* User recovery
+
+---
+
+# CLI Backup Commands
+
+## Create Full Backup
+
+```bash
+bzod backup
+```
+
+Output:
+
+```text
+backups/
+└── backup-YYYYMMDD-HHMMSS.zip
+```
+
+---
+
+## Create User Backup
+
+```bash
+bzod backup-user 42
+```
+
+Output:
+
+```text
+backups/
+└── user-42-YYYYMMDD-HHMMSS.zip
+```
+
+---
+
+# CLI Restore Commands
+
+## Restore Full Backup
+
+```bash
+bzod restore backup-20260619-020000.zip
+```
+
+Restores:
+
+* users.db
+* system.db
+* all tenant databases
+
+---
+
+## Restore Single User
+
+```bash
+bzod restore-user user-42-20260619.zip
+```
+
+Restores only:
+
+```text
+users/42/content.db
+users/42/analytics.db
+```
+
+without affecting any other tenant.
+
+---
+
+# Web-Based Backup Management
+
+Administrative users can manage backups through:
+
+```text
+/admin/backups
+```
+
+Features:
+
+* Create backup
+* Download backup
+* Upload backup
+* Restore backup
+* Delete backup
+
+Only authenticated administrators may access backup operations.
+
+---
+
+# Backup Strategy
+
+## Recommended Schedule
+
+### Daily
+
+```text
+02:00 AM
+```
+
+Create a full platform backup.
+
+---
+
+### Weekly
+
+```text
+Sunday 03:00 AM
+```
+
+Create a full backup and copy it to:
+
+* NAS
+* Secondary server
+* External storage
+
+---
+
+### Monthly
+
+Archive a backup for long-term retention.
+
+Recommended retention:
+
+```text
+12 months
+```
+
+---
+
+# Retention Policy
+
+Recommended policy:
+
+```text
+Daily Backups:
+30 days
+
+Weekly Backups:
+12 weeks
+
+Monthly Backups:
+12 months
+```
+
+Adjust retention according to compliance requirements.
+
+---
+
+# Upgrade Procedure
+
+Always create a backup before upgrading.
+
+## Step 1
+
+Create backup:
+
+```bash
+bzod backup
+```
+
+## Step 2
+
+Upgrade BZOD binary.
+
+## Step 3
+
+Start BZOD.
+
+```bash
+bzod serve
+```
+
+## Step 4
+
+Allow database migrations to complete.
+
+## Step 5
+
+Verify:
+
+* Login
+* URLs
+* Landing pages
+* Analytics
+* Administration panels
+
+---
+
+# Restore Validation
+
+After every restore operation verify:
+
+## Authentication
+
+* Administrator login works
+* Standard user login works
+
+## Content
+
+* URLs are visible
+* Landing pages render correctly
+
+## Routing
+
+* Slug redirects work
+* Landing page routes resolve
+
+## Analytics
+
+* Visit counts exist
+* Analytics dashboards load
+
+## System
+
+* Audit events visible
+* Moderation records preserved
+* System settings preserved
+
+## Multi-User
+
+* Tenant isolation maintained
+* Ownership mappings preserved
+
+---
+
+# Disaster Recovery Scenarios
+
+## Scenario 1: Deleted User
+
+Problem:
+
+```text
+User account accidentally deleted.
+```
+
+Recovery:
+
+```bash
+bzod restore-user user-42.zip
+```
+
+Verify:
+
+* URLs restored
+* Pages restored
+* Analytics restored
+
+---
+
+## Scenario 2: Corrupted Tenant Database
+
+Problem:
+
+```text
+content.db corruption
+```
+
+Recovery:
+
+```bash
+bzod restore-user user-42.zip
+```
+
+or
+
+```bash
+bzod restore full-backup.zip
+```
+
+---
+
+## Scenario 3: Corrupted users.db
+
+Problem:
+
+```text
+Unable to login
+Missing users
+Session failures
+```
+
+Recovery:
+
+```bash
+bzod restore full-backup.zip
+```
+
+---
+
+## Scenario 4: Corrupted system.db
+
+Problem:
+
+```text
+Slug resolution failures
+Moderation data missing
+Settings lost
+```
+
+Recovery:
+
+```bash
+bzod restore full-backup.zip
+```
+
+---
+
+## Scenario 5: Complete Server Failure
+
+Problem:
+
+```text
+Disk failure
+Server loss
+Hardware replacement
+```
+
+Recovery:
+
+1. Reinstall operating system
+2. Install BZOD
+3. Restore backup
+
+```bash
+bzod restore backup.zip
+```
+
+4. Start BZOD
+
+```bash
+bzod serve
+```
+
+---
+
+# WAL Mode
+
+BZOD uses SQLite Write-Ahead Logging (WAL).
+
+Examples:
+
+```text
+users.db
+users.db-wal
+users.db-shm
+
+system.db
+system.db-wal
+system.db-shm
+
+content.db
+content.db-wal
+content.db-shm
+
+analytics.db
+analytics.db-wal
+analytics.db-shm
+```
+
+Benefits:
+
+* Improved concurrency
+* Better crash recovery
+* Faster write operations
+
+---
+
+# Backup Safety
+
+Do not manually copy live SQLite databases while the server is actively writing.
+
+Always use:
+
+```bash
+bzod backup
+```
+
+or the Backup Management UI.
+
+This ensures consistent snapshots.
+
+---
+
+# Security Considerations
+
+Backups may contain:
+
+* User accounts
+* Password hashes
+* Session metadata
+* Analytics data
+* Audit records
+* API token hashes
+
+Even though passwords and tokens are stored as hashes, backup archives should be treated as sensitive information.
+
+Recommended practices:
+
+* Encrypt backup storage
+* Restrict filesystem permissions
+* Maintain offsite copies
+* Transfer backups over secure channels
+* Test restores periodically
+
+---
+
+# Backup Testing
+
+A backup is only useful if it can be restored.
+
+Quarterly validation is recommended.
+
+Example:
+
+```bash
+mkdir restore-test
+
+bzod restore backup.zip \
+    --data-dir restore-test
+```
+
+Verify:
+
+* Login works
+* URLs resolve
+* Landing pages load
+* Analytics display
+* Administration dashboard functions
+
+---
+
+# Production Recommendation
+
+Minimum production policy:
+
+```text
+Daily Full Backup
+Weekly Offsite Backup
+Monthly Archive Backup
+Quarterly Restore Validation
+```
+
+Following this policy protects against:
+
+* User mistakes
+* Database corruption
+* Upgrade failures
+* Hardware failures
+* Site disasters
+
+and provides a reliable recovery path for BZOD deployments.
@@ -0,0 +1,363 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on Keep a Changelog and this project follows Semantic Versioning.
+
+---
+
+# v0.6.0 — Legacy Restore Compatibility & Version Reporting
+
+- **Legacy Backup Restore**: Full backward-compatible restore support for `legacy_flat_backup` archives into the current multi-tenant database architecture
+- **CLI Version Reporting**: Added `--version` / `-V` flags derived from Cargo package metadata
+- **Deploy Script**: Removed obsolete `init-db` command; database creation and migration now handled by `bzod serve`
+- **Version Verification**: Deploy script now verifies installed binary version matches requested version
+
+# v0.5.3 — Architecture Refinement & Redirect Hardening
+
+---
+
+## Changed
+
+### Architecture
+
+* Eliminated the monolithic `admin.rs` handler file
+* Reorganized admin functionality into focused feature modules under `src/web/admin/`
+* Separated authentication, dashboard, URLs, pages, analytics, settings, users, sessions, quotas, health, backups, API keys, audit, and moderation into dedicated modules
+* Extracted shared authentication and authorization helpers
+* Extracted common export and helper functionality
+
+### Redirect Handling
+
+* Removed panic-prone `HeaderValue::from_str(...).unwrap()` pattern from the redirect path
+* Added destination URL validation (scheme validation, control character rejection)
+* Added safe HTTP Location header construction
+* Improved database error logging with structured fields
+* Reduced unnecessary database mutex lock acquisitions on the redirect hot path
+* Removed synchronous expiration writes from the redirect hot path
+
+---
+
+## Improved
+
+* Database lock scoping across admin handlers
+* Error handling consistency and observability
+* Handler decomposition for oversized functions
+* Reduced duplicated handler logic across admin operations
+
+---
+
+## Verified
+
+* Root landing page (GET /) confirmed as intentional route serving www/index.html
+* Release binary built successfully
+* Runtime smoke tests passed (GET /, GET /login, GET /admin/login all return HTTP 200)
+* SQLite WAL mode and foreign-key enforcement initialized successfully
+* All existing migrations reported as up to date
+* Comprehensive automated test suite passed, including:
+  * Authentication and migration tests
+  * Redirect security tests
+  * Root landing page test
+  * Backup and restore tests
+  * Business workflow tests
+  * Security tests
+  * Slug namespace, registry, and transfer tests
+  * User management and isolation tests
+  * WAL recovery tests
+  * HTTP end-to-end tests
+
+---
+
+## Notes
+
+* This release is an internal architecture and quality improvement
+* No new user-facing features were introduced
+* Existing API and route behavior was preserved
+* Existing redirect security and tenant isolation behavior was preserved
+
+---
+
+# v0.5.1 - General Availability (GA)
+
+Release Date: 2026-06-20
+
+BZOD v0.5.1 is the largest release since project inception, transforming BZOD from a single-user URL shortener into a complete multi-user redirector, landing page, analytics, and administration platform.
+
+---
+
+## Added
+
+### Multi-User Platform
+
+* Multi-user architecture with isolated tenant databases
+* Standard user accounts
+* Administrator accounts
+* User provisioning and lifecycle management
+* User enable/disable operations
+* User deletion workflows
+* Password reset functionality
+* User quota management
+* User database isolation
+
+### Authentication & Security
+
+* Session-based authentication
+* CSRF protection
+* Role-Based Access Control (RBAC)
+* Password hashing and verification
+* Session invalidation
+* Login/logout workflows
+* Administrative privilege separation
+* Audit logging
+
+### User Self-Service Portal
+
+* User dashboard
+* My Links management
+* My Pages management
+* User analytics dashboard
+* API token management
+* Password management
+* Profile management
+
+### Administration
+
+* User management dashboard
+* User detail pages
+* User creation forms
+* User editing interface
+* Session administration
+* Quota administration
+* Moderation dashboard
+* Slug management dashboard
+* Audit event viewer
+* Backup management interface
+* System health dashboard
+
+### Analytics
+
+* Per-user analytics
+* URL analytics dashboards
+* Landing page analytics dashboards
+* Browser statistics
+* Referrer tracking
+* Visit logging
+* Geographic analytics framework
+* Analytics aggregation jobs
+
+### Content Management
+
+* Landing page builder
+* URL registry management
+* Global slug namespace
+* Slug ownership tracking
+* Slug transfer workflows
+* Soft delete support
+* Moderation controls
+
+### Operations
+
+* Backup CLI
+* Restore CLI
+* User backup support
+* User restore support
+* Database diagnostics
+* Health checks
+* Quota reconciliation jobs
+* Retention jobs
+* Expiry jobs
+* Aggregation workers
+
+### Documentation
+
+* Installation Guide
+* Upgrade Guide
+* Multi-User Guide
+* Administration Guide
+* Security Guide
+* Backup & Restore Guide
+* Database Documentation
+* Architecture Documentation
+* CLI Documentation
+* API Documentation
+* Testing Documentation
+
+---
+
+## Changed
+
+### Architecture
+
+* Migrated from single-user storage model to tenant-isolated storage model
+* Introduced users.db as central identity store
+* Introduced system.db as global platform metadata store
+* Introduced per-user content databases
+* Introduced per-user analytics databases
+
+### Routing
+
+* Unified global slug resolution
+* Centralized slug ownership tracking
+* Improved redirect handling
+* Improved landing page routing
+
+### Analytics
+
+* Improved aggregation performance
+* Improved reporting consistency
+* Improved analytics isolation
+
+### Administration
+
+* Expanded administrative tooling
+* Improved dashboard coverage
+* Added operational visibility
+
+---
+
+## Security
+
+### Added
+
+* CSRF validation
+* Session management
+* RBAC enforcement
+* Audit event logging
+* User isolation controls
+* Slug ownership validation
+
+### Hardened
+
+* Authentication flows
+* Session validation
+* Administrative authorization
+* User lifecycle operations
+
+---
+
+## Database
+
+### Added
+
+* users.db
+* system.db
+* Per-user content.db
+* Per-user analytics.db
+* Migration framework
+
+### Improved
+
+* WAL mode support
+* Upgrade migrations
+* Backup compatibility
+* Recovery workflows
+
+---
+
+## Testing
+
+### Added
+
+Comprehensive automated validation covering:
+
+* Authentication tests
+* Authorization tests
+* Migration tests
+* Upgrade validation tests
+* User isolation tests
+* Slug namespace tests
+* Slug transfer tests
+* Moderation tests
+* Backup and restore tests
+* Disaster recovery tests
+* Analytics tests
+* Concurrency tests
+* HTTP end-to-end tests
+* Business workflow tests
+* Security regression tests
+
+### Coverage
+
+* 90+ unit and integration tests
+* HTTP workflow validation
+* Upgrade path verification
+* Multi-user isolation verification
+* Backup and recovery validation
+
+---
+
+## Fixed
+
+### Authentication
+
+* Multi-user migration login regressions
+* Session validation issues
+* Administrative account migration edge cases
+
+### Routing
+
+* Redirect handling consistency
+* Slug ownership synchronization
+* Landing page resolution issues
+
+### Analytics
+
+* Aggregation edge cases
+* Reporting consistency
+* Isolation validation
+
+### Concurrency
+
+* Fixed mutex deadlock conditions discovered during E2E testing
+* Improved lock scoping around audit logging
+
+### Administration
+
+* Improved slug transfer workflows
+* Improved user lifecycle operations
+* Improved dashboard consistency
+
+---
+
+## Upgrade Notes
+
+### From v0.4.0
+
+BZOD v0.5.0 introduces a new multi-user architecture.
+
+Existing installations are automatically migrated during startup.
+
+Migration includes:
+
+* Legacy administrator migration
+* Global slug index generation
+* User database creation
+* Analytics preservation
+* Content preservation
+
+Backups are strongly recommended before upgrading.
+
+---
+
+# v0.4.0
+
+## Added
+
+* Raw visitor activity logs
+* Analytics drill-down pages
+* Date-range analytics filters
+* CSV export
+* JSON export
+* Advanced pagination
+* Visitor log tables
+
+## Improved
+
+* Registry pagination
+* Analytics navigation
+* Export performance
+
+## Fixed
+
+* Pagination edge cases
+* Analytics sorting consistency
@@ -0,0 +1,286 @@
+# BZOD Command Line Interface (CLI)
+
+BZOD includes a comprehensive command-line interface for server administration, backups, migrations, diagnostics, validation, and multi-user management.
+
+The current command list for BZOD v0.6.0 is:
+
+```text
+$ bzod --help
+
+BZOD - Personal Redirector & Landing Page Platform
+
+Usage: bzod <COMMAND>
+
+Commands:
+  serve           Start the BZOD web server
+  backup          Create a tar.gz backup of all databases
+  restore         Restore databases from a tar.gz backup file
+  migrate         Apply pending database schema migrations
+  stats           Print database statistics and record counts in the terminal
+  validate        Perform a one-shot validation of all registered short link destinations
+  create-admin    Create a new administrator user in the database
+  doctor          Run database diagnostics and health checks
+  shorten         Shorten a URL (Feature 3)
+  expand          Expand a shortened code or custom slug to its destination URL (Feature 4)
+  create-user     Create a new standard user in the database
+  delete-user     Delete a standard user and all their databases/slugs
+  disable-user    Disable a standard user
+  enable-user     Enable a standard user
+  reset-password  Reset standard user's password
+  list-users      List all standard/system users
+  backup-user     Backup a standard user's databases to a .tar.zst package
+  restore-user    Restore a standard user's databases from a .tar.zst package
+  admin-migrate   FUTURE: Migrate legacy admin content to a specific admin tenant database
+  repair          Repair registry and database inconsistencies
+  help            Print this message or the help of the given subcommand(s)
+
+Options:
+  -h, --help  Print help
+```
+
+---
+
+# Server Operations
+
+## Start Web Server
+
+```bash
+bzod serve
+```
+
+---
+
+# Backup & Recovery
+
+## Full Backup
+
+```bash
+bzod backup
+```
+
+Creates a compressed backup archive containing:
+
+* users.db
+* system.db
+* content databases
+* analytics databases
+* user directories
+
+## Full Restore
+
+```bash
+bzod restore backup.tar.gz
+```
+
+Restores an entire BZOD installation from a backup archive.
+
+---
+
+# Database Operations
+
+## Apply Migrations
+
+```bash
+bzod migrate
+```
+
+Applies any pending database migrations.
+
+Safe to execute multiple times.
+
+## Database Statistics
+
+```bash
+bzod stats
+```
+
+Displays database statistics, record counts, storage usage, and operational metrics.
+
+---
+
+# Validation & Diagnostics
+
+## Validate Links
+
+```bash
+bzod validate
+```
+
+Checks all registered URLs and reports invalid destinations.
+
+## Health Diagnostics
+
+```bash
+bzod doctor
+```
+
+Performs:
+
+* SQLite integrity checks
+* WAL validation
+* Database availability checks
+* Storage verification
+* System health diagnostics
+* Global registry integrity validation
+
+## Registry Repair
+
+```bash
+bzod repair registry --dry-run
+```
+
+Provides a transaction-safe repair utility for fixing global slug registry inconsistencies detected by `bzod doctor`.
+
+* Use `--dry-run` to preview changes safely.
+* Use `--force` to execute changes and remove orphaned entries.
+* Use `--slug <slug>` to target a single missing entry.
+
+---
+
+# URL Management
+
+## Create Short URL
+
+```bash
+bzod shorten https://example.com
+```
+
+## Expand Existing URL
+
+```bash
+bzod expand abc123
+```
+
+Returns the destination URL associated with the slug.
+
+---
+
+# Administrator Management
+
+## Create Administrator
+
+```bash
+bzod create-admin admin
+```
+
+Creates a new administrator account.
+
+---
+
+# User Management
+
+## List Users
+
+```bash
+bzod list-users
+```
+
+Displays all users in the platform.
+
+## Create User
+
+```bash
+bzod create-user alice
+```
+
+Creates a new standard user.
+
+## Disable User
+
+```bash
+bzod disable-user alice
+```
+
+Blocks login and invalidates sessions.
+
+## Enable User
+
+```bash
+bzod enable-user alice
+```
+
+Re-enables a disabled user.
+
+## Reset Password
+
+```bash
+bzod reset-password alice
+```
+
+Resets a user's password.
+
+## Delete User
+
+```bash
+bzod delete-user alice
+```
+
+Deletes:
+
+* User account
+* User databases
+* Sessions
+* API tokens
+* Slug ownership
+
+---
+
+# User Backup Operations
+
+## Backup User
+
+```bash
+bzod backup-user alice
+```
+
+Creates a portable `.tar.zst` archive containing all user-owned data.
+
+## Restore User
+
+```bash
+bzod restore-user alice.tar.zst
+```
+
+Restores a user from a previously generated archive.
+
+---
+
+# Recommended Maintenance
+
+Daily:
+
+```bash
+bzod doctor
+```
+
+Weekly:
+
+```bash
+bzod backup
+```
+
+Before Upgrades:
+
+```bash
+bzod backup
+bzod validate
+```
+
+After Upgrades:
+
+```bash
+bzod migrate
+bzod doctor
+```
+
+---
+
+# Related Documentation
+
+* INSTALL.md
+* MULTI_USER.md
+* ADMIN_GUIDE.md
+* BACKUP_RESTORE.md
+* SECURITY.md
+* API.md
+* ARCHITECTURE.md
@@ -0,0 +1,354 @@
+# BZOD v0.6.0 vs Self-Hosted URL Management Platforms
+
+BZOD is a modern, privacy-focused, self-hosted URL Management Platform written in Rust and developed as part of the NX9 Platform.
+
+Unlike traditional URL shorteners that focus primarily on URL redirection, BZOD provides a complete platform for managing URLs, landing pages, analytics, users, permissions, backups, and operational workflows.
+
+## Quick Comparison
+
+| Feature                  | BZOD | Shlink | YOURLS | Chhoto URL |
+|--------------------------|------|--------|--------|------------|
+| Language                 | Rust | PHP    | PHP    | Rust       |
+| Single Binary            | ✅    | ❌      | ❌      | ✅          |
+| Landing Pages            | ✅    | ❌      | Plugin | ❌          |
+| QR Code + Analytics      | ✅    | Partial| Plugin | Partial    |
+| Password Protection      | ✅    | Limited| Plugin | ❌          |
+| Backup & Restore         | ✅    | External| External| ❌         |
+| Audit Trail              | ✅    | Limited| Plugin | ❌          |
+| CLI Tools                | ✅    | Limited| Limited| Limited    |
+| Dependencies             | None | PHP + DB | PHP + DB | None     |
+| Deployment Complexity    | Low  | Medium | High   | Low        |
+
+---
+### Rust URL Shortener Comparison
+| Project              | Language | Single Binary | Landing Pages | QR Codes + Analytics | Password Protection | Backup & Restore | CLI Tools   | Audit Trail | Admin Dashboard | Notes                                      |
+|----------------------|----------|---------------|---------------|----------------------|---------------------|------------------|-------------|-------------|-----------------|--------------------------------------------|
+| **BZOD**             | Rust     | ✅ (~11 MB)    | ✅             | ✅                    | ✅                   | ✅                | ✅           | ✅           | ✅               | Feature-rich, multi-user ready, strong philosophy |
+| Chhoto URL           | Rust     | ✅             | ❌             | Partial              | ❌                   | ❌                | Limited     | ❌           | Basic           | Very minimal, smallest footprint           |
+| smrs                 | Rust     | ✅             | ❌             | ❌                    | ❌                   | ❌                | Limited     | ❌           | Basic           | Personal project, very simple              |
+| urlshortener-rs      | Rust     | Library       | N/A           | N/A                  | N/A                  | N/A               | N/A         | N/A          | N/A             | Library, not full server                   |
+| Custom Rust          | Rust     | Varies        | Varies        | Varies               | Varies               | Varies            | Varies      | Varies       | Varies          | Usually minimal implementations            |
+
+# Executive Summary
+
+BZOD combines:
+
+* URL shortening
+* Landing pages
+* QR code generation
+* QR analytics
+* Link analytics
+* Password-protected links
+* Link expiration
+* REST API
+* Administrative dashboard
+* Multi-user operation
+* User management
+* User quotas
+* Session management
+* Audit logging
+* Moderation
+* Backup & restore
+* Disaster recovery tooling
+
+into a single Rust binary deployment.
+
+---
+
+# At a Glance
+
+| Feature              | BZOD              |
+| -------------------- | ----------------- |
+| Language             | Rust              |
+| License              | MIT OR Apache-2.0 |
+| Deployment           | Single Binary     |
+| Runtime Dependencies | None              |
+| Database             | SQLite            |
+| Multi-User           | Yes               |
+| Landing Pages        | Yes               |
+| QR Codes             | Yes               |
+| Analytics            | Yes               |
+| REST API             | Yes               |
+| CLI Tools            | Yes               |
+| Backups              | Built-in          |
+| Audit Logs           | Built-in          |
+| RBAC                 | Built-in          |
+
+---
+
+# What Changed in v0.5.0
+
+BZOD v0.5.0 introduces a major architectural evolution.
+
+## New Platform Capabilities
+
+* Multi-user architecture
+* Tenant isolation
+* Global slug namespace
+* User management
+* User quotas
+* Session management
+* Administrative dashboards
+* User self-service dashboards
+* Audit event logging
+* Moderation workflows
+* Backup management
+* Health monitoring
+* Upgrade framework
+* Migration tooling
+
+BZOD is no longer merely a URL shortener.
+
+It is now a self-hosted URL Management Platform.
+
+---
+
+# Traditional URL Shortener Comparison
+
+| Capability          | BZOD | Shlink   | YOURLS   | Chhoto URL |
+| ------------------- | ---- | -------- | -------- | ---------- |
+| URL Shortening      | ✅    | ✅        | ✅        | ✅          |
+| Landing Pages       | ✅    | ❌        | Plugin   | ❌          |
+| QR Generation       | ✅    | Partial  | Plugin   | Partial    |
+| QR Analytics        | ✅    | Partial  | Plugin   | ❌          |
+| Password Protection | ✅    | Limited  | Plugin   | ❌          |
+| Link Expiration     | ✅    | ✅        | Plugin   | Limited    |
+| REST API            | ✅    | ✅        | ✅        | JSON-RPC   |
+| Backup & Restore    | ✅    | External | External | ❌          |
+| Audit Logs          | ✅    | Limited  | Plugin   | ❌          |
+| Multi User          | ✅    | Partial  | Plugin   | ❌          |
+| User Quotas         | ✅    | ❌        | ❌        | ❌          |
+| User Isolation      | ✅    | ❌        | ❌        | ❌          |
+| User Dashboards     | ✅    | ❌        | ❌        | ❌          |
+
+---
+
+# Multi-User Platform Comparison
+
+BZOD v0.5.0 introduces first-class multi-user support.
+
+| Capability             | BZOD |
+| ---------------------- | ---- |
+| User Accounts          | ✅    |
+| Administrator Accounts | ✅    |
+| User Isolation         | ✅    |
+| User Quotas            | ✅    |
+| Session Management     | ✅    |
+| API Tokens             | ✅    |
+| Audit Trail            | ✅    |
+| Moderation             | ✅    |
+| Tenant Analytics       | ✅    |
+| Self-Service Portal    | ✅    |
+
+Most self-hosted URL shorteners are fundamentally single-user applications.
+
+BZOD is designed for:
+
+* Individuals
+* Teams
+* Organizations
+* Educational Institutions
+* Governments
+* Service Providers
+
+---
+
+# Security Comparison
+
+| Security Feature          | BZOD | Typical URL Shortener |
+| ------------------------- | ---- | --------------------- |
+| Argon2id Password Hashing | ✅    | Varies                |
+| Session Management        | ✅    | Basic                 |
+| CSRF Protection           | ✅    | Varies                |
+| RBAC                      | ✅    | Rare                  |
+| Audit Logging             | ✅    | Rare                  |
+| User Disablement          | ✅    | Rare                  |
+| Moderation Controls       | ✅    | Rare                  |
+| Tenant Isolation          | ✅    | Rare                  |
+| API Token Security        | ✅    | Varies                |
+
+---
+
+# Operations Comparison
+
+| Operational Feature | BZOD |
+| ------------------- | ---- |
+| Backup Creation     | ✅    |
+| Backup Restore      | ✅    |
+| User Backup         | ✅    |
+| User Restore        | ✅    |
+| Disaster Recovery   | ✅    |
+| Upgrade Validation  | ✅    |
+| Health Monitoring   | ✅    |
+| WAL Recovery        | ✅    |
+| Migration Framework | ✅    |
+
+Most competing products rely on external tooling for these capabilities.
+
+---
+
+# Deployment Comparison
+
+| Requirement                | BZOD | Shlink   | YOURLS   |
+| -------------------------- | ---- | -------- | -------- |
+| Single Binary              | ✅    | ❌        | ❌        |
+| SQLite Only                | ✅    | Optional | Optional |
+| External Database Required | ❌    | Usually  | Usually  |
+| Docker Support             | ✅    | ✅        | ✅        |
+| Systemd Support            | ✅    | Manual   | Manual   |
+| Backup Framework           | ✅    | ❌        | ❌        |
+| Upgrade Framework          | ✅    | ❌        | ❌        |
+
+---
+
+# BZOD vs Go-Based URL Shorteners
+
+Popular Go alternatives include:
+
+* Krtk
+* Goshorly
+* Slash
+* Shortr
+* Custom Gin/Echo implementations
+
+### Strengths of Go Projects
+
+* Small binaries
+* Excellent performance
+* Simple codebases
+
+### Strengths of BZOD
+
+* Multi-user support
+* Landing pages
+* User management
+* Built-in analytics
+* Backup framework
+* Audit logging
+* Moderation
+* Administrative dashboards
+
+---
+
+# BZOD vs Python-Based Solutions
+
+Examples:
+
+* Pygmy
+* Schort
+* ReducePy
+* Flask-based projects
+* FastAPI-based projects
+
+### Python Advantages
+
+* Rapid development
+* Familiar ecosystem
+
+### BZOD Advantages
+
+* No runtime dependency
+* Lower memory consumption
+* Single binary deployment
+* Operational tooling included
+* Better long-term maintenance characteristics
+
+---
+
+# Reliability & Testing
+
+BZOD v0.5.0 includes a comprehensive automated validation suite.
+
+Coverage includes:
+
+* Unit tests
+* Integration tests
+* HTTP E2E tests
+* Business workflow tests
+* Upgrade validation tests
+* Backup/restore tests
+* Disaster recovery tests
+* Security tests
+* Concurrency tests
+* WAL recovery tests
+
+The platform is validated using more than 90 automated tests.
+
+---
+
+# NX9 Platform Philosophy
+
+BZOD follows the NX9 engineering philosophy:
+
+* Linux-first
+* Rust-first
+* Self-hosted
+* Privacy-first
+* No telemetry
+* No vendor lock-in
+* No external dependencies
+* Single binary deployment
+
+The goal is simple:
+
+> Build software that remains useful, understandable, maintainable, and deployable decades into the future.
+
+---
+
+# Who Should Use BZOD?
+
+BZOD is suitable for:
+
+### Individuals
+
+* Personal URL management
+* Homelabs
+* Self-hosted services
+
+### Organizations
+
+* Marketing campaigns
+* Internal redirects
+* Landing page hosting
+
+### Governments
+
+* Public service redirects
+* Long-term link preservation
+* Controlled infrastructure
+
+### Service Providers
+
+* Multi-tenant URL management
+* Managed short-link services
+* White-label deployments
+
+---
+
+# Conclusion
+
+BZOD v0.5.0 is not simply a URL shortener.
+
+It is a self-hosted URL Management Platform providing:
+
+* Multi-user operation
+* Tenant isolation
+* URL shortening
+* Landing pages
+* QR generation
+* Analytics
+* Audit logging
+* Moderation
+* User administration
+* Backup & restore
+* Health monitoring
+
+within a single Rust binary deployment.
+
+BZOD is designed for individuals, organizations, governments, educational institutions, and service providers that require full ownership of their links, analytics, and infrastructure.
+
+> Own your links.
+> Own your data.
+> Own your infrastructure.
+
+No telemetry. No vendor lock-in. No unnecessary complexity.
@@ -0,0 +1,503 @@
+# DATABASES.md
+
+# BZOD Database Architecture
+
+BZOD v0.6.0 uses SQLite exclusively.
+
+Rather than using a single monolithic database, BZOD separates data into administrative and tenant-specific databases. This architecture improves security, isolation, backup flexibility, disaster recovery, and scalability.
+
+---
+
+# Overview
+
+BZOD stores data in the following structure:
+
+```text
+data/
+├── admin/
+│   ├── admin.db
+│   ├── system.db
+│   └── users.db
+│
+└── users/
+    ├── 1/
+    │   ├── analytics.db
+    │   ├── content.db
+    │   └── profile.db
+    │
+    ├── 2/
+    │   ├── analytics.db
+    │   ├── content.db
+    │   └── profile.db
+    │
+    └── N/
+        ├── analytics.db
+        ├── content.db
+        └── profile.db
+```
+
+Each user receives isolated databases.
+
+No user content or analytics are stored in the central administrative databases.
+
+---
+
+# Administrative Databases
+
+Administrative databases are located under:
+
+```text
+data/admin/
+```
+
+---
+
+# users.db
+
+Primary authentication and user management database.
+
+Purpose:
+
+* User accounts
+* Password hashes
+* Sessions
+* Quotas
+* API tokens
+* User status tracking
+
+Typical tables:
+
+```text
+users
+sessions
+quotas
+api_tokens
+```
+
+Responsibilities:
+
+* Authentication
+* Authorization
+* Session management
+* Account status
+* Quota enforcement
+
+This is the primary identity database of the platform.
+
+---
+
+# system.db
+
+Global platform database.
+
+Purpose:
+
+* Global slug namespace
+* Moderation
+* Auditing
+* System configuration
+
+Typical tables:
+
+```text
+global_slugs
+slug_history
+moderation_events
+audit_events
+reserved_slugs
+settings
+```
+
+Responsibilities:
+
+* Global slug uniqueness
+* Slug ownership
+* Moderation actions
+* Audit logging
+* System settings
+
+Every redirect ultimately resolves through records stored in this database.
+
+---
+
+# admin.db
+
+Administrative application database.
+
+Purpose:
+
+* Administrative metadata
+* Administrative API key records
+* Legacy compatibility structures
+* Internal management data
+
+Typical tables:
+
+```text
+api_keys
+audit_events
+```
+
+This database is reserved for administrative functions and does not store tenant content.
+
+---
+
+# Tenant Databases
+
+Tenant databases are located under:
+
+```text
+data/users/{user_id}/
+```
+
+Each user owns a completely isolated set of databases.
+
+Example:
+
+```text
+data/users/2/
+├── analytics.db
+├── content.db
+└── profile.db
+```
+
+---
+
+# content.db
+
+Stores user-owned content.
+
+Purpose:
+
+* Short URLs
+* Landing pages
+* QR metadata
+* Preview metadata
+
+Typical tables:
+
+```text
+urls
+pages
+qr_codes
+previews
+```
+
+Responsibilities:
+
+* URL management
+* Landing page management
+* Content ownership
+
+This database contains the actual resources owned by a user.
+
+---
+
+# analytics.db
+
+Stores traffic and visitor information.
+
+Purpose:
+
+* Visit recording
+* Referrer tracking
+* Browser tracking
+* Country statistics
+* Aggregated analytics
+
+Typical tables:
+
+```text
+visits
+referrers
+browsers
+countries
+daily_stats
+```
+
+Responsibilities:
+
+* Analytics collection
+* Reporting
+* Dashboard statistics
+
+Analytics are fully isolated per user.
+
+Administrators access aggregated analytics by querying each user's analytics database.
+
+---
+
+# profile.db
+
+Stores user-specific profile information.
+
+Purpose:
+
+* User preferences
+* Profile settings
+* Future extensible metadata
+
+Typical tables:
+
+```text
+profile
+preferences
+```
+
+Responsibilities:
+
+* User profile management
+* Dashboard preferences
+* Future personalization features
+
+---
+
+# Database Isolation Model
+
+BZOD follows a strict tenant isolation model.
+
+```text
+User A
+ ├── content.db
+ ├── analytics.db
+ └── profile.db
+
+User B
+ ├── content.db
+ ├── analytics.db
+ └── profile.db
+```
+
+User databases never share tables.
+
+Cross-user content access is prevented by design.
+
+Benefits:
+
+* Security
+* Easier backups
+* Easier deletion
+* Reduced corruption impact
+
+---
+
+# Global Slug Registry
+
+The system maintains a single namespace.
+
+Stored in:
+
+```text
+system.db
+```
+
+Table:
+
+```text
+global_slugs
+```
+
+Example:
+
+```text
+abc123 → User 2 URL
+docs → User 5 Page
+demo → User 1 URL
+```
+
+This guarantees:
+
+* Global uniqueness
+* Ownership tracking
+* Moderation support
+* Slug transfer support
+
+---
+
+# Write Flow
+
+Creating a URL:
+
+```text
+1. Validate quota
+2. Register slug in system.db
+3. Create URL in content.db
+4. Update quota counters
+5. Write audit event
+```
+
+Creating a landing page:
+
+```text
+1. Validate quota
+2. Register slug in system.db
+3. Create page in content.db
+4. Update quota counters
+5. Write audit event
+```
+
+---
+
+# Analytics Flow
+
+Visitor request:
+
+```text
+GET /abc123
+```
+
+Process:
+
+```text
+global_slugs
+        ↓
+content.db lookup
+        ↓
+redirect
+        ↓
+analytics.db visit record
+```
+
+Analytics writes never modify content records.
+
+---
+
+# WAL Mode
+
+All databases operate in SQLite WAL mode.
+
+Verify:
+
+```sql
+PRAGMA journal_mode;
+```
+
+Expected:
+
+```text
+wal
+```
+
+Benefits:
+
+* Improved concurrency
+* Reduced write contention
+* Crash recovery
+
+Associated files:
+
+```text
+*.db
+*.db-shm
+*.db-wal
+```
+
+---
+
+# WAL Checkpointing
+
+Large WAL files are normal during heavy traffic.
+
+Example:
+
+```text
+analytics.db-wal
+content.db-wal
+```
+
+To manually checkpoint:
+
+```sql
+PRAGMA wal_checkpoint(TRUNCATE);
+```
+
+The healthcheck and backup jobs may trigger checkpoints automatically.
+
+---
+
+# Backups
+
+Recommended:
+
+```bash
+bzod backup
+```
+
+This creates a consistent archive of:
+
+```text
+admin/
+users/
+```
+
+Never manually copy live databases while the application is running.
+
+---
+
+# Integrity Verification
+
+Run:
+
+```bash
+bzod doctor
+```
+
+Or:
+
+```sql
+PRAGMA integrity_check;
+```
+
+Expected:
+
+```text
+ok
+```
+
+---
+
+# Migration System
+
+BZOD maintains schema versions using:
+
+```sql
+PRAGMA user_version;
+```
+
+Startup automatically executes:
+
+```text
+Db::init()
+```
+
+which:
+
+1. Creates missing databases
+2. Applies migrations
+3. Validates schemas
+4. Repairs legacy installations when required
+
+---
+
+# Design Principles
+
+BZOD database architecture prioritizes:
+
+* SQLite-only deployment
+* Multi-user isolation
+* Operational simplicity
+* Backup friendliness
+* Easy disaster recovery
+* Minimal dependencies
+* Single-binary deployment
+
+---
+
+# Related Documentation
+
+* ARCHITECTURE.md
+* MULTI_USER.md
+* BACKUP_RESTORE.md
+* INSTALL.md
+* UPGRADE.md
+* SECURITY.md
@@ -0,0 +1,545 @@
+# Docker Deployment Guide for BZOD
+
+This guide covers deployment, upgrades, backup, restore, troubleshooting, and production best practices for **BZOD (nx9-url-shortener)** using Docker.
+
+---
+
+# Overview
+
+BZOD is a lightweight self-hosted URL shortener and landing page platform written in Rust.
+
+Features include:
+
+* URL shortening
+* Human-readable custom slugs (`!office`, `!home`, etc.)
+* Landing pages
+* QR code generation
+* Analytics
+* Audit logging
+* API access
+* Backup and restore
+* SQLite-based storage
+* Docker deployment
+
+BZOD is designed to remain simple:
+
+* No PostgreSQL
+* No Redis
+* No external dependencies
+* No vendor lock-in
+
+---
+
+# Quick Start
+
+## Clone Repository
+
+```bash
+git clone https://github.com/thakares/nx9-url-shortener.git
+cd nx9-url-shortener
+```
+
+## Build and Start
+
+```bash
+docker compose up -d --build
+```
+
+## Create Administrator
+
+```bash
+docker exec -it bzod bzod create-admin
+```
+
+Open:
+
+```text
+http://SERVER-IP:8654
+```
+
+Admin panel:
+
+```text
+http://SERVER-IP:8654/admin
+```
+
+---
+
+# Docker Compose
+
+Example:
+
+```yaml
+services:
+  bzod:
+    container_name: bzod
+    build: .
+    restart: unless-stopped
+
+    ports:
+      - "8654:8654"
+
+    volumes:
+      - ./data:/app/data
+      - ./config:/app/config
+
+    environment:
+      HOST: 0.0.0.0
+      PORT: 8654
+      DATA_DIR: /app/data
+      COOKIE_SECURE: "false"
+
+    healthcheck:
+      test: ["CMD", "./bzod", "doctor"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+```
+
+Start:
+
+```bash
+docker compose up -d
+```
+
+Verify:
+
+```bash
+docker ps
+docker logs -f bzod
+```
+
+---
+
+# Directory Layout
+
+Typical deployment:
+
+```text
+bzod/
+├── docker-compose.yml
+├── Dockerfile
+├── config/
+├── data/
+│   ├── admin.db
+│   ├── content.db
+│   ├── analytics.db
+│   └── system.db
+└── backups/
+```
+
+---
+
+# Root Landing Page
+
+BZOD can serve a static landing page from:
+
+```text
+www/index.html
+```
+
+This page is available at:
+
+```text
+https://your-domain/
+```
+
+Examples:
+
+```text
+https://bzo.in/
+https://short.example.com/
+```
+
+The root landing page is packaged automatically inside the Docker image.
+
+---
+
+# Reverse Proxy Configuration
+
+BZOD is intended to run behind a reverse proxy.
+
+Example Nginx configuration:
+
+```nginx
+server {
+    server_name bzo.in;
+
+    location / {
+        proxy_pass http://127.0.0.1:8654;
+
+        proxy_set_header Host $host;
+        proxy_set_header X-Real-IP $remote_addr;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+        proxy_set_header X-Forwarded-Proto $scheme;
+    }
+}
+```
+
+Example deployment:
+
+```text
+Internet
+    ↓
+Nginx Proxy Manager
+    ↓
+BZOD Docker Container
+```
+
+---
+
+# Environment Variables
+
+| Variable      | Description        | Default   |
+| ------------- | ------------------ | --------- |
+| HOST          | Bind address       | 0.0.0.0   |
+| PORT          | Listen port        | 8654      |
+| DATA_DIR      | Database directory | /app/data |
+| COOKIE_SECURE | Secure cookies     | false     |
+| RUST_LOG      | Logging level      | info      |
+
+Production recommendation:
+
+```text
+COOKIE_SECURE=true
+```
+
+when HTTPS is enabled.
+
+---
+# Analytics Export
+
+Analytics pages support:
+
+* Raw visitor logs
+* CSV export
+* JSON export
+* Date filtering
+
+Exports can be generated from:
+
+Admin → Analytics
+
+Backup
+---
+
+## Web UI
+
+Navigate to:
+
+```text
+Admin → Settings → Maintenance & DB Utilities
+```
+
+Click:
+
+```text
+Download Backup
+```
+
+A compressed archive containing all databases will be downloaded.
+
+---
+
+## CLI Backup
+
+Create backup:
+
+```bash
+docker exec -it bzod bzod backup
+```
+
+Example output:
+
+```text
+backup-2026-06-14.tar.gz
+```
+
+---
+
+# Restore
+
+## Web UI Restore
+
+Navigate to:
+
+```text
+Admin → Settings → Maintenance & DB Utilities
+```
+
+Upload:
+
+```text
+backup.tar.gz
+```
+
+Type:
+
+```text
+RESTORE
+```
+
+Confirm restore.
+
+The system will:
+
+1. Validate archive contents
+2. Restore databases
+3. Reinitialize database access
+4. Redirect to login
+
+---
+
+## CLI Restore
+
+Copy backup archive into container or mounted volume.
+
+Run:
+
+```bash
+docker exec -it bzod bash
+
+cd /app/data
+
+bzod restore --file backup.tar.gz
+```
+
+---
+
+# Disaster Recovery
+
+Example recovery procedure:
+
+```bash
+docker compose down
+
+# Restore backup archive
+
+docker compose up -d
+```
+
+Verify:
+
+```bash
+docker exec -it bzod bzod doctor
+docker exec -it bzod bzod validate
+```
+
+Check:
+
+* URLs
+* Landing pages
+* Analytics
+* Audit logs
+* Settings
+
+---
+
+# Useful CLI Commands
+
+Health:
+
+```bash
+docker exec -it bzod bzod doctor
+```
+
+Statistics:
+
+```bash
+docker exec -it bzod bzod stats
+```
+
+Validate databases:
+
+```bash
+docker exec -it bzod bzod validate
+```
+
+Create admin:
+
+```bash
+docker exec -it bzod bzod create-admin
+```
+
+Shorten URL:
+
+```bash
+docker exec -it bzod bzod shorten https://example.com
+```
+
+Custom slug:
+
+```bash
+docker exec -it bzod bzod shorten https://example.com --slug !office
+```
+
+Expand URL:
+
+```bash
+docker exec -it bzod bzod expand !office
+```
+
+---
+
+# Upgrading
+
+Pull latest source:
+
+```bash
+git pull
+```
+
+Rebuild:
+
+```bash
+docker compose build --no-cache
+```
+
+Restart:
+
+```bash
+docker compose up -d
+```
+
+Verify:
+
+```bash
+docker logs -f bzod
+```
+# Upgrading to v0.4.0
+
+1. Backup databases
+2. Pull latest source
+3. Rebuild container
+4. Restart service
+
+```bash
+git pull
+docker compose build --no-cache
+docker compose up -d
+---
+
+# Troubleshooting
+
+## Read-Only SQLite Database
+
+Symptoms:
+
+```text
+attempt to write a readonly database
+```
+
+Check ownership:
+
+```bash
+ls -lah data/
+```
+
+Fix permissions:
+
+```bash
+docker exec -u 0 -it bzod bash
+
+chown -R bzod:bzod /app/data
+```
+
+docker exec -it bzod bzod doctor
+
+Restart:
+
+```bash
+docker compose restart bzod
+```
+
+---
+
+## Missing Root Landing Page
+
+Symptoms:
+
+```text
+404 on /
+```
+
+Verify:
+
+```bash
+docker exec -it bzod ls -lah /app/www
+```
+
+Expected:
+
+```text
+/app/www/index.html
+```
+
+Rebuild image if necessary:
+
+```bash
+docker compose build --no-cache
+docker compose up -d
+```
+
+---
+
+## Health Check Failure
+
+Inspect logs:
+
+```bash
+docker logs bzod
+```
+
+Run:
+
+```bash
+docker exec -it bzod bzod doctor
+```
+
+---
+
+## Port Already In Use
+
+Change host port mapping:
+
+```yaml
+ports:
+  - "8080:8654"
+```
+
+Access:
+
+```text
+http://SERVER-IP:8080
+```
+
+---
+
+# Production Recommendations
+
+* Use HTTPS
+* Run behind Nginx Proxy Manager or Nginx
+* Use strong administrator credentials
+* Schedule regular backups
+* Periodically test restore procedures
+* Monitor disk space
+* Keep Docker images updated
+
+---
+
+# Validation Checklist
+
+After deployment verify:
+
+* [ ] Admin login works
+* [ ] URL shortening works
+* [ ] Custom slugs work
+* [ ] Landing pages work
+* [ ] QR generation works
+* [ ] Analytics recorded
+* [ ] Backup download works
+* [ ] Restore workflow works
+* [ ] Root landing page loads
+* [ ] `bzod doctor` reports healthy
+
+A deployment should not be considered production-ready until backup and restore procedures have been successfully tested.
@@ -0,0 +1,604 @@
+# BZOD Installation Guide
+
+Version: v0.6.0
+
+---
+
+# Introduction
+
+BZOD is a self-hosted multi-user URL management platform written in Rust.
+
+Features include:
+
+* URL shortening
+* Landing pages
+* QR code generation
+* Analytics
+* User management
+* Audit logging
+* Moderation
+* Backup & restore
+* Disaster recovery
+
+BZOD is distributed as a single executable and uses SQLite databases for storage.
+
+No PostgreSQL, MySQL, Redis, Elasticsearch, or external services are required.
+
+---
+
+# Installation Methods
+
+BZOD supports three deployment methods:
+
+| Method         | Recommended For  |
+| -------------- | ---------------- |
+| Docker Compose | Most deployments |
+| Native Binary  | Linux servers    |
+| Source Build   | Development      |
+
+---
+
+# System Requirements
+
+## Minimum
+
+| Component | Requirement  |
+| --------- | ------------ |
+| CPU       | 1 Core       |
+| Memory    | 512 MB       |
+| Storage   | 1 GB         |
+| OS        | Linux x86_64 |
+
+## Recommended
+
+| Component | Requirement              |
+| --------- | ------------------------ |
+| CPU       | 2+ Cores                 |
+| Memory    | 2 GB                     |
+| Storage   | 10+ GB SSD               |
+| OS        | Debian 12 / Ubuntu 24.04 |
+
+## Tested Platforms
+
+* Debian 12 Bookworm
+* Ubuntu 22.04
+* Ubuntu 24.04
+* Arch Linux
+* Docker
+* CasaOS
+
+---
+
+# Installation Using Docker
+
+## Prerequisites
+
+Install:
+
+```bash
+docker
+docker compose
+```
+
+Verify:
+
+```bash
+docker --version
+docker compose version
+```
+
+---
+
+## Create Directory
+
+```bash
+mkdir -p /opt/bzod
+cd /opt/bzod
+```
+
+---
+
+## Copy Files
+
+Required:
+
+```text
+docker-compose.yml
+Dockerfile
+```
+
+Optional:
+
+```text
+bzod.service
+```
+
+---
+
+## Start Container
+
+```bash
+docker compose up -d
+```
+
+Verify:
+
+```bash
+docker compose ps
+```
+
+View logs:
+
+```bash
+docker compose logs -f
+```
+
+---
+
+## Stop Container
+
+```bash
+docker compose down
+```
+
+---
+
+## Restart Container
+
+```bash
+docker compose restart
+```
+
+---
+
+# Native Installation
+
+## Install Dependencies
+
+### Debian / Ubuntu
+
+```bash
+sudo apt update
+
+sudo apt install -y \
+    build-essential \
+    pkg-config \
+    libssl-dev \
+    sqlite3
+```
+
+### Arch Linux
+
+```bash
+sudo pacman -S \
+    base-devel \
+    openssl \
+    sqlite
+```
+
+---
+
+## Download Release Binary
+
+Example:
+
+```bash
+wget https://example.com/bzod-v0.6.0-linux-amd64.tar.gz
+```
+
+Extract:
+
+```bash
+tar -xzf bzod-v0.6.0-linux-amd64.tar.gz
+```
+
+Install:
+
+```bash
+sudo install -m755 bzod /usr/local/bin/bzod
+```
+
+Verify:
+
+```bash
+bzod --help
+```
+
+---
+
+# Build From Source
+
+## Install Rust
+
+```bash
+curl https://sh.rustup.rs -sSf | sh
+```
+
+Verify:
+
+```bash
+cargo --version
+rustc --version
+```
+
+---
+
+## Clone Repository
+
+```bash
+git clone https://github.com/thakares/nx9-url-shortener.git
+
+cd nx9-url-shortener
+```
+
+---
+
+## Build
+
+Development:
+
+```bash
+cargo build
+```
+
+Release:
+
+```bash
+cargo build --release
+```
+
+Binary:
+
+```bash
+target/release/bzod
+```
+
+---
+
+# Data Directory
+
+BZOD automatically creates its databases on first startup.
+
+Default structure:
+
+```text
+data/
+├── users.db
+├── system.db
+│
+├── admin/
+│   ├── content.db
+│   └── analytics.db
+│
+└── users/
+    └── ...
+```
+
+Do not manually modify database files while BZOD is running.
+
+---
+
+# First Startup
+
+Run:
+
+```bash
+bzod serve
+```
+
+By default:
+
+```text
+http://localhost:8080
+```
+
+Open:
+
+```text
+http://localhost:8080
+```
+
+---
+
+# Bootstrap Administrator
+
+On a fresh installation:
+
+1. Open Login page
+2. Use bootstrap credentials
+3. Create the first administrator account
+4. Save the credentials securely
+
+After bootstrap:
+
+* Bootstrap mode is disabled
+* Normal authentication is enforced
+
+---
+
+# Create Administrator Using CLI
+
+Alternative method:
+
+```bash
+bzod create-admin
+```
+
+Follow prompts:
+
+```text
+Username:
+Password:
+```
+
+The administrator account is stored in:
+
+```text
+users.db
+```
+
+---
+
+# Reverse Proxy Configuration
+
+Using Nginx is recommended.
+
+Example:
+
+```nginx
+server {
+    server_name bzod.example.com;
+
+    location / {
+        proxy_pass http://127.0.0.1:8080;
+
+        proxy_set_header Host $host;
+        proxy_set_header X-Real-IP $remote_addr;
+
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+
+        proxy_set_header X-Forwarded-Proto $scheme;
+    }
+}
+```
+
+Reload:
+
+```bash
+sudo nginx -t
+sudo systemctl reload nginx
+```
+
+---
+
+# HTTPS
+
+Recommended options:
+
+* Let's Encrypt
+* Nginx Proxy Manager
+* Caddy
+* Traefik
+
+Always use HTTPS in production.
+
+---
+
+# Running as Systemd Service
+
+Install binary:
+
+```bash
+sudo install -m755 bzod /usr/local/bin/bzod
+```
+
+Copy service:
+
+```bash
+sudo cp bzod.service /etc/systemd/system/
+```
+
+Reload:
+
+```bash
+sudo systemctl daemon-reload
+```
+
+Enable:
+
+```bash
+sudo systemctl enable bzod
+```
+
+Start:
+
+```bash
+sudo systemctl start bzod
+```
+
+Status:
+
+```bash
+sudo systemctl status bzod
+```
+
+Logs:
+
+```bash
+journalctl -u bzod -f
+```
+
+---
+
+# Firewall
+
+Open HTTP:
+
+```bash
+sudo ufw allow 8080/tcp
+```
+
+HTTPS:
+
+```bash
+sudo ufw allow 443/tcp
+```
+
+HTTP:
+
+```bash
+sudo ufw allow 80/tcp
+```
+
+---
+
+# Health Verification
+
+Open:
+
+```text
+http://localhost:8080
+```
+
+Login as administrator.
+
+Verify:
+
+* Dashboard loads
+* User list loads
+* URL creation works
+* Landing pages work
+* QR generation works
+* Analytics record visits
+
+---
+
+# Upgrade Procedure
+
+Always backup before upgrading.
+
+Create backup:
+
+```bash
+bzod backup
+```
+
+Stop service:
+
+```bash
+sudo systemctl stop bzod
+```
+
+Replace binary.
+
+Run migrations:
+
+```bash
+bzod migrate
+```
+
+Start service:
+
+```bash
+sudo systemctl start bzod
+```
+
+Verify logs.
+
+See:
+
+```text
+docs/UPGRADE.md
+```
+
+---
+
+# Troubleshooting
+
+## Port Already In Use
+
+Check:
+
+```bash
+ss -tulpn | grep 8080
+```
+
+Change port or stop conflicting service.
+
+---
+
+## Database Locked
+
+Verify only one BZOD instance is running:
+
+```bash
+ps aux | grep bzod
+```
+
+---
+
+## Permission Errors
+
+Verify ownership:
+
+```bash
+chown -R bzod:bzod data/
+```
+
+---
+
+## Login Problems
+
+Verify:
+
+* Administrator account exists
+* Session cookies enabled
+* System clock is correct
+
+---
+
+## View Logs
+
+Systemd:
+
+```bash
+journalctl -u bzod -f
+```
+
+Docker:
+
+```bash
+docker compose logs -f
+```
+
+---
+
+# Next Steps
+
+After installation:
+
+1. Read `MULTI_USER.md`
+2. Read `ADMIN_GUIDE.md`
+3. Configure backups
+4. Configure HTTPS
+5. Create additional users
+6. Verify restore procedures
+
+---
+
+# Additional Documentation
+
+| File              | Purpose                  |
+| ----------------- | ------------------------ |
+| ARCHITECTURE.md   | System architecture      |
+| MULTI_USER.md     | Multi-user design        |
+| ADMIN_GUIDE.md    | Administrative workflows |
+| BACKUP_RESTORE.md | Backup procedures        |
+| SECURITY.md       | Security model           |
+| CLI.md            | Command reference        |
+| API.md            | REST API reference       |
+| UPGRADE.md        | Upgrade instructions     |
+
+---
+
+End of Document.
@@ -0,0 +1,732 @@
+# BZOD Multi-User Architecture Guide
+
+Version: v0.6.0
+
+---
+
+# Introduction
+
+BZOD v0.5.0 introduces a complete multi-user architecture that transforms BZOD from a single-tenant URL shortener into a secure, isolated, self-hosted multi-user platform.
+
+Each user receives logically isolated content and analytics storage while sharing a common authentication, administration, moderation, and routing infrastructure.
+
+This document explains the architecture, database layout, ownership model, security boundaries, quotas, slug management, and administrative workflows.
+
+---
+
+# Design Goals
+
+The multi-user architecture was designed around the following principles:
+
+* Strong tenant isolation
+* Single binary deployment
+* SQLite-only operation
+* Minimal operational complexity
+* No external services required
+* Global slug namespace
+* Centralized administration
+* Disaster recovery support
+* Simple backup and restore workflows
+
+---
+
+# User Types
+
+BZOD supports the following account types.
+
+## Administrator
+
+Administrators can:
+
+* Access the administrative dashboard
+* Create users
+* Delete users
+* Reset passwords
+* Manage quotas
+* Transfer ownership
+* Moderate content
+* Manage backups
+* Access health dashboards
+* Access audit logs
+
+Administrators cannot bypass database isolation.
+
+---
+
+## Standard User
+
+Standard users can:
+
+* Create short URLs
+* Create landing pages
+* View analytics
+* Generate QR codes
+* Manage API tokens
+* Update passwords
+
+Standard users cannot:
+
+* Access other user content
+* Access administrative functions
+* Access system settings
+
+---
+
+## System Accounts
+
+System accounts are reserved for internal operations.
+
+They cannot authenticate into the dashboard.
+
+---
+
+# Database Architecture
+
+BZOD uses multiple SQLite databases.
+
+## users.db
+
+Central identity store.
+
+Contains:
+
+```text
+users
+sessions
+quotas
+api_tokens
+```
+
+Responsibilities:
+
+* Authentication
+* Session management
+* Password verification
+* User status management
+* Quota tracking
+
+---
+
+## system.db
+
+Global platform database.
+
+Contains:
+
+```text
+global_slugs
+slug_history
+moderation_events
+audit_events
+settings
+reserved_slugs
+```
+
+Responsibilities:
+
+* Slug ownership
+* Moderation
+* Audit logging
+* Global settings
+* System metadata
+
+---
+
+## Tenant Databases
+
+Every tenant owns independent databases.
+
+Example:
+
+```text
+users/
+└── 15/
+    ├── content.db
+    └── analytics.db
+```
+
+Responsibilities:
+
+### content.db
+
+Stores:
+
+```text
+urls
+pages
+qr_metadata
+previews
+```
+
+### analytics.db
+
+Stores:
+
+```text
+visits
+aggregates
+referrers
+browsers
+countries
+```
+
+---
+
+# Directory Structure
+
+Example installation:
+
+```text
+data/
+├── users.db
+├── system.db
+│
+├── admin/
+│   ├── content.db
+│   └── analytics.db
+│
+└── users/
+    ├── 2/
+    │   ├── content.db
+    │   └── analytics.db
+    │
+    ├── 3/
+    │   ├── content.db
+    │   └── analytics.db
+    │
+    └── 4/
+        ├── content.db
+        └── analytics.db
+```
+
+---
+
+# Global Slug Namespace
+
+BZOD uses a platform-wide namespace.
+
+A slug can only exist once.
+
+Examples:
+
+```text
+/company
+/about
+/docs
+```
+
+If User A owns:
+
+```text
+/company
+```
+
+User B cannot create:
+
+```text
+/company
+```
+
+The operation is rejected.
+
+---
+
+# Slug Registration Flow
+
+When a URL or page is created:
+
+1. Validate quota.
+2. Validate slug.
+3. Register slug in system.db.
+4. Create record in tenant content.db.
+5. Increment quota counters.
+6. Write audit log.
+
+If any step fails:
+
+* Changes are rolled back.
+* Partial records are removed.
+
+---
+
+# Global Slug Table
+
+Conceptually:
+
+```text
+global_slugs
+```
+
+Contains:
+
+```text
+slug
+owner_user_id
+target_type
+target_id
+status
+created_at
+```
+
+Example:
+
+| slug | owner | type |
+| ---- | ----- | ---- |
+| docs | 3     | page |
+| api  | 8     | page |
+| home | 2     | url  |
+
+---
+
+# Slug Ownership Transfer
+
+Administrators may transfer ownership.
+
+Process:
+
+1. Validate destination quotas.
+2. Copy content.
+3. Move ownership.
+4. Update global slug registry.
+5. Record history.
+6. Write audit event.
+
+Analytics remain preserved.
+
+URLs remain functional.
+
+---
+
+# Tenant Isolation
+
+Each user owns independent databases.
+
+Example:
+
+```text
+User A
+└── users/2/
+
+User B
+└── users/3/
+```
+
+User A never accesses:
+
+```text
+users/3/content.db
+users/3/analytics.db
+```
+
+User B never accesses:
+
+```text
+users/2/content.db
+users/2/analytics.db
+```
+
+All access is enforced by application logic.
+
+---
+
+# Authentication Architecture
+
+Authentication is centralized.
+
+Stored in:
+
+```text
+users.db
+```
+
+Tables:
+
+```text
+users
+sessions
+```
+
+All dashboard sessions use:
+
+```text
+bzod_session
+```
+
+Sessions are validated against:
+
+```text
+users.db.sessions
+```
+
+---
+
+# Session Lifecycle
+
+Login:
+
+```text
+User Login
+    ↓
+Create Session
+    ↓
+Store in users.db
+    ↓
+Set bzod_session cookie
+```
+
+Logout:
+
+```text
+Delete session row
+    ↓
+Expire cookie
+```
+
+Disabled users immediately lose access.
+
+---
+
+# Quota System
+
+Every user has quotas.
+
+Examples:
+
+```text
+max_urls
+max_pages
+max_storage_mb
+max_api_tokens
+```
+
+Current utilization is tracked separately.
+
+Administrators may:
+
+* Increase limits
+* Reduce limits
+* Trigger reconciliation
+
+---
+
+# Quota Reconciliation
+
+Background job:
+
+```text
+quota_reconcile
+```
+
+Purpose:
+
+* Detect drift
+* Recount resources
+* Repair counters
+
+Example:
+
+```text
+Stored URLs = 50
+Actual URLs = 47
+```
+
+Counter automatically corrected.
+
+---
+
+# Analytics Isolation
+
+Each tenant stores analytics independently.
+
+Example:
+
+```text
+users/10/analytics.db
+```
+
+Contains only User 10 traffic.
+
+Administrators can:
+
+* View aggregated analytics
+* Access user analytics
+
+Users cannot view analytics from other tenants.
+
+---
+
+# QR Code System
+
+QR codes are generated dynamically.
+
+Endpoints:
+
+```text
+/api/qr/{slug}.png
+/api/qr/{slug}.svg
+```
+
+Slug ownership is resolved through:
+
+```text
+system.db.global_slugs
+```
+
+No content database scan is required.
+
+---
+
+# Moderation Architecture
+
+Administrators can:
+
+* Flag content
+* Disable content
+* Delete content
+* Transfer ownership
+
+Disabled content returns:
+
+```http
+410 Gone
+```
+
+For:
+
+```text
+/slug
+/p/slug
+/api/qr/slug.png
+/api/qr/slug.svg
+```
+
+---
+
+# Audit Logging
+
+All administrative actions are recorded.
+
+Examples:
+
+```text
+login
+logout
+user_create
+user_delete
+password_reset
+quota_update
+slug_transfer
+backup_create
+restore_execute
+```
+
+Stored in:
+
+```text
+system.db
+```
+
+---
+
+# Backup Architecture
+
+Supported levels:
+
+## Full Platform Backup
+
+Includes:
+
+```text
+users.db
+system.db
+all tenant databases
+```
+
+---
+
+## User Backup
+
+Includes:
+
+```text
+content.db
+analytics.db
+```
+
+For a specific user.
+
+---
+
+# Disaster Recovery
+
+Supported operations:
+
+```bash
+bzod backup
+bzod restore
+bzod backup-user
+bzod restore-user
+```
+
+Recovery preserves:
+
+* URLs
+* Pages
+* Analytics
+* Users
+* Slugs
+* Settings
+
+---
+
+# Upgrade Path
+
+BZOD automatically migrates:
+
+```text
+v0.4.x
+```
+
+to
+
+```text
+v0.5.x
+```
+
+Migration process:
+
+1. Create users.db.
+2. Create system.db.
+3. Create admin tenant.
+4. Migrate content.
+5. Migrate analytics.
+6. Populate global_slugs.
+7. Create legacy_admin.
+8. Validate integrity.
+
+No manual database migration is normally required.
+
+---
+
+# Security Model
+
+Security boundaries:
+
+## Authentication
+
+Centralized.
+
+```text
+users.db
+```
+
+---
+
+## Authorization
+
+Role-based.
+
+```text
+admin
+standard
+system
+```
+
+---
+
+## CSRF Protection
+
+All forms protected.
+
+Invalid tokens:
+
+```http
+403 Forbidden
+```
+
+---
+
+## Session Security
+
+* Secure session IDs
+* Session invalidation
+* Expiration support
+* Replay protection
+
+---
+
+## Tenant Isolation
+
+Per-user databases.
+
+No shared content tables.
+
+---
+
+# Operational Recommendations
+
+Recommended deployment:
+
+```text
+Nginx
+    ↓
+BZOD
+    ↓
+SQLite WAL
+```
+
+Enable:
+
+* HTTPS
+* Daily backups
+* Log rotation
+* Health monitoring
+
+---
+
+# Limitations
+
+Current v0.5.0 limitations:
+
+* SQLite backend only
+* Single server deployment
+* No clustering
+* No federation
+* No organization account hierarchy
+
+These may be addressed in future releases.
+
+---
+
+# Future Expansion
+
+Potential v0.6.x features:
+
+* Organization accounts
+* Service accounts
+* SSO integration
+* Multi-node replication
+* Advanced analytics dashboards
+* Scheduled tasks UI
+
+---
+
+# Summary
+
+BZOD v0.5.0 provides:
+
+* Centralized authentication
+* Multi-user isolation
+* Global slug namespace
+* Per-user analytics
+* Administrative moderation
+* Quotas
+* Audit logging
+* Backup & disaster recovery
+* Single-binary deployment
+
+while remaining lightweight, SQLite-native, and operationally simple.
+
+---
+
+End of Document.
@@ -0,0 +1,419 @@
+# BZOD v0.6.0 — Legacy Restore Compatibility & Version Reporting
+
+Release Date: 2026-08-09
+
+## Highlights
+
+- **Legacy Backup Restore Compatibility**: Backups created with the web admin "Download Backup" feature (`legacy_flat_backup` format) can now be correctly restored into the current multi-tenant database architecture. Previously, these restores failed with "no such table: users" because the restore validator ran against the empty legacy `users.db` before layout normalization.
+
+- **CLI Version Reporting**: `bzod --version` and `bzod -V` now report the application version derived from Cargo.toml package metadata, ensuring the reported version cannot diverge from the build.
+
+- **Deploy Script Modernization**: Removed the obsolete `init-db` command from the deployment script. Database creation and schema migration are now handled automatically by `bzod serve`. The deploy script now verifies the installed binary version using `--version`.
+
+## Breaking Changes
+
+None.
+
+# BZOD v0.5.3 — Architecture Refinement & Redirect Hardening
+
+BZOD v0.5.3 is an internal quality and maintainability release focused on architectural refinement, redirect handler hardening, and comprehensive verification.
+
+No new user-facing features are introduced. Existing API contracts, route behavior, authentication, and tenant isolation are fully preserved.
+
+---
+
+# Highlights
+
+## Modular Admin Architecture
+
+The former monolithic admin handler file was eliminated and replaced with a focused module directory at `src/web/admin/`.
+
+Feature modules:
+
+* `auth.rs` — authentication and session handling
+* `dashboard.rs` — dashboard rendering
+* `urls.rs` — URL management handlers
+* `pages.rs` — landing page management handlers
+* `analytics.rs` — analytics and export handlers
+* `settings.rs` — settings and configuration handlers
+* `users.rs` — user management handlers
+* `sessions.rs` — session administration
+* `quotas.rs` — quota management
+* `health.rs` — health diagnostics
+* `backups.rs` — backup and restore handlers
+* `api_keys.rs` — API key management
+* `audit.rs` — audit log handlers
+* `moderation.rs` — content moderation handlers
+
+Benefits:
+
+* Improved code organization and navigability
+* Reduced coupling between feature areas
+* Improved database lock scoping
+* Reduced duplicated handler logic
+* Better error handling consistency and observability
+* Simplified future extension
+
+---
+
+## Redirect Handler Hardening
+
+The public redirect path (`GET /:code`) was hardened against invalid HTTP Location header values.
+
+Changes:
+
+* Removed the panic-prone `HeaderValue::from_str(...).unwrap()` pattern
+* Added destination URL validation (scheme enforcement, control character rejection)
+* Added safe Location header construction that handles malformed values gracefully
+* Improved database error logging with structured fields
+* Reduced unnecessary database mutex lock acquisitions
+* Removed synchronous expiration writes from the redirect hot path
+
+Existing redirect security and tenant isolation behavior was preserved.
+
+---
+
+## Root Landing Page Verification
+
+* Confirmed `GET /` as an intentional application route serving `www/index.html`
+* Resolved a runtime path-resolution issue affecting static landing-page resolution
+* Verified `GET /` returns HTTP 200
+* Verified `GET /login` returns HTTP 200
+* Verified `GET /admin/login` returns HTTP 200
+
+---
+
+# Testing & Validation
+
+BZOD v0.5.3 passed:
+
+* Release build (`cargo build --release`)
+* Comprehensive automated test suite, including:
+  * Authentication and migration tests
+  * Redirect security tests
+  * Root landing page test
+  * Backup and restore tests
+  * Business workflow tests
+  * Security tests
+  * Slug namespace, registry, and transfer tests
+  * User management and isolation tests
+  * WAL recovery tests
+  * HTTP end-to-end tests
+* Runtime smoke tests against the release binary
+* SQLite WAL mode and foreign-key enforcement initialization
+* Database migration verification (all migrations up to date)
+
+---
+
+# Compatibility
+
+* No breaking changes
+* No API changes
+* No route changes
+* No database schema changes
+* No configuration changes
+* Direct upgrade from v0.5.1 with no migration required
+
+---
+
+# Repository
+
+* Clean source tree established
+* Build artifacts, temporary reports, and IDE metadata removed
+* Existing BZOD Git history preserved
+* Refactoring baseline merged with existing history
+
+---
+
+---
+
+# BZOD v0.5.1 — Namespace Integrity & Platform Hardening
+
+**Release Date:** 2026-06-20
+
+BZOD v0.5.1 focuses on platform integrity, multi-tenant safety, dashboard parity, QR reliability, and upgrade validation.
+
+While v0.5.0 introduced the multi-user architecture, v0.5.1 strengthens the foundations required for safe operation at scale.
+
+---
+
+# Highlights
+## Runtime Efficiency (v0.5.1)
+
+| Metric              | Value      |
+|---------------------|------------|
+| Binary Size         | 11 MB      |
+| RSS Memory          | 11.8 MB    |
+| Peak RSS            | 11.8 MB    |
+| CPU Idle            | 0.02%      |
+| Swap Usage          | 0 KB       |
+| PIDs                | 7          |
+
+**On a typical 32 GB server:**
+- Memory usage: ~0.04%
+- No swapping
+- Plenty of headroom
+
+BZOD runs closer to a lightweight infrastructure service than a typical web application.
+
+## Global Slug Registry
+
+Introduced a hardened global slug registry to guarantee namespace integrity across the entire platform.
+
+The following resources can no longer share the same slug:
+
+* Administrator URLs
+* Administrator Landing Pages
+* User URLs
+* User Landing Pages
+
+Duplicate namespace conflicts are automatically detected and blocked.
+
+---
+
+## Namespace Integrity Validation
+
+New validation routines now verify:
+
+* Duplicate slug detection
+* Missing ownership records
+* Invalid registry entries
+* Invalid target types
+* Orphaned slug references
+
+Namespace conflicts now abort upgrades and restores before corruption can occur.
+
+---
+
+## Reservation-Based Slug Allocation
+
+BZOD now reserves slugs before content creation.
+
+Creation workflow:
+
+```text
+Quota Check
+↓
+Reserve Global Slug
+↓
+Create Content
+↓
+Activate Slug
+↓
+Increment Quota
+↓
+Audit Log
+```
+
+Benefits:
+
+* Prevents race conditions
+* Prevents duplicate creation under concurrency
+* Enables safer rollback handling
+
+---
+
+## Stale Reservation Recovery
+
+Added automatic cleanup of abandoned slug reservations.
+
+Scenarios covered:
+
+* Server crash during creation
+* Interrupted writes
+* Failed transactions
+
+BZOD now automatically recovers stale reservations during startup.
+
+---
+
+## Dashboard Parity
+
+Administrator and Standard User dashboards now provide equivalent functionality where appropriate.
+
+Added parity validation for:
+
+* URL management
+* Landing page management
+* Analytics
+* QR code previews
+* Export functionality
+
+Differences remain only for administrator-specific operations.
+
+---
+
+## Unified Analytics Templates
+
+Removed duplicated analytics templates.
+
+Benefits:
+
+* Consistent rendering
+* Reduced maintenance burden
+* Improved reliability
+
+Administrator and user analytics now share the same rendering logic.
+
+---
+
+## QR Code Improvements
+
+QR functionality was substantially improved.
+
+### Added
+
+* Inline QR previews
+* PNG downloads
+* SVG downloads
+* Shared QR rendering component
+
+### Fixed
+
+* Landing page QR generation
+* Multi-user QR ownership handling
+* QR routing consistency
+* Content-type validation
+
+---
+
+## Canonical Landing Page Routing
+
+Landing page slugs now redirect permanently to canonical page URLs.
+
+Example:
+
+```text
+/landing-page
+```
+
+redirects to:
+
+```text
+/p/landing-page
+```
+
+using:
+
+```http
+301 Moved Permanently
+```
+
+This improves consistency and SEO behavior.
+
+---
+
+## Ownership Isolation Hardening
+
+Additional protections ensure:
+
+* Users cannot access another user's analytics
+* Users cannot export another user's data
+* Users cannot manage another user's resources
+
+New ownership validation tests were added.
+
+---
+
+## Backup & Restore Improvements
+
+Restore operations now validate namespace integrity before importing data.
+
+Benefits:
+
+* No silent slug collisions
+* No partial restores
+* No hidden ownership conflicts
+
+Restore operations fail safely when conflicts are detected.
+
+---
+
+## Upgrade Validation Enhancements
+
+Upgrade workflows now verify:
+
+* Global namespace consistency
+* Duplicate slug conflicts
+* Registry integrity
+* Tenant ownership correctness
+
+Unsafe upgrades are blocked automatically.
+
+---
+
+## Health & Diagnostics
+
+The system health subsystem now validates:
+
+* Global slug registry integrity
+* Namespace conflicts
+* Ownership consistency
+* Stale reservations
+
+This improves operational visibility and troubleshooting.
+
+---
+
+# Testing & Validation
+
+BZOD v0.5.1 passed:
+
+* Formatting validation (`cargo fmt --check`)
+* Static analysis (`cargo clippy --all-targets -- -D warnings`)
+* Full automated test suite
+* Namespace integrity tests
+* Ownership isolation tests
+* QR endpoint tests
+* Dashboard parity tests
+* Upgrade validation tests
+* Backup & restore tests
+* Disaster recovery tests
+* Security tests
+* Concurrency tests
+
+All automated tests pass successfully.
+
+---
+
+# Upgrade Notes
+
+Administrators upgrading from v0.5.0 should review:
+
+* UPGRADE.md
+* MULTI_USER.md
+* BACKUP_RESTORE.md
+* DATABASES.md
+* TESTING.md
+
+BZOD will automatically validate namespace integrity before completing upgrades.
+
+Duplicate slugs that previously existed across users or resource types must be resolved before migration can proceed.
+
+---
+
+# Breaking Changes
+
+## Global Namespace Enforcement
+
+Slugs are now globally unique across the entire platform.
+
+Configurations that previously relied on duplicate slugs across users or resource types will be rejected during upgrade.
+
+This behavior is intentional and protects routing integrity.
+
+---
+
+# Summary
+
+BZOD v0.5.1 is an integrity-focused release that significantly strengthens:
+
+* Namespace safety
+* Multi-tenant isolation
+* Dashboard consistency
+* QR reliability
+* Restore safety
+* Upgrade safety
+* Operational diagnostics
+
+The result is a more predictable, recoverable, and production-ready platform.
@@ -0,0 +1,685 @@
+# BZOD Security Guide
+
+Version: v0.6.0
+
+---
+
+# Security Overview
+
+BZOD is designed as a self-hosted URL shortener and landing page platform with a strong emphasis on:
+
+* Multi-user isolation
+* Secure authentication
+* Role-based access control
+* Auditability
+* Data ownership
+* Disaster recovery
+* Operational simplicity
+
+This document describes the security architecture, threat model, authentication mechanisms, authorization controls, and operational security recommendations for BZOD v0.6.0.
+
+---
+
+# Security Principles
+
+BZOD follows several core principles:
+
+1. Least Privilege
+2. Tenant Isolation
+3. Defense in Depth
+4. Auditability
+5. Secure Defaults
+6. Explicit Ownership
+7. Fail Secure
+
+---
+
+# Threat Model
+
+BZOD is designed to protect against:
+
+* Unauthorized dashboard access
+* Credential theft
+* Session hijacking
+* Cross-user data access
+* Slug takeover attempts
+* Privilege escalation
+* CSRF attacks
+* XSS injection attempts
+* Unauthorized API access
+* Malicious content modification
+* Accidental administrative mistakes
+
+BZOD is not intended to defend against:
+
+* Physical server compromise
+* Root-level operating system compromise
+* Malware running as the BZOD service user
+* Full database theft by a privileged host administrator
+
+---
+
+# Authentication
+
+Authentication is centralized in:
+
+```text
+users.db
+```
+
+Tables:
+
+```text
+users
+sessions
+api_tokens
+```
+
+All users authenticate through the same identity system.
+
+---
+
+# Password Security
+
+Passwords are never stored in plaintext.
+
+Stored values:
+
+```text
+password_hash
+```
+
+Passwords are hashed before storage.
+
+Administrative password resets generate entirely new hashes.
+
+Existing passwords cannot be recovered.
+
+---
+
+# Session Security
+
+All dashboard authentication uses:
+
+```text
+bzod_session
+```
+
+cookie.
+
+Sessions are stored in:
+
+```text
+users.db.sessions
+```
+
+Each session contains:
+
+```text
+session_id
+user_id
+created_at
+expires_at
+```
+
+---
+
+## Session Validation
+
+Each authenticated request verifies:
+
+1. Session exists
+2. Session has not expired
+3. User exists
+4. User status is active
+5. User has required permissions
+
+Failure at any step immediately invalidates access.
+
+---
+
+## Session Revocation
+
+Sessions are revoked when:
+
+* User logs out
+* User is disabled
+* User is deleted
+* Password is reset
+* Administrator revokes sessions
+
+---
+
+## Session Fixation Protection
+
+BZOD generates new session identifiers after successful authentication.
+
+Previously issued identifiers are not reused.
+
+---
+
+# Authorization Model
+
+BZOD implements Role-Based Access Control (RBAC).
+
+Supported roles:
+
+```text
+admin
+standard
+system
+```
+
+---
+
+## Administrator
+
+Administrators can:
+
+* Manage users
+* Reset passwords
+* Transfer ownership
+* Manage quotas
+* Access audit logs
+* Review analytics
+* Create backups
+* Restore backups
+* Moderate content
+
+Administrators cannot bypass audit logging.
+
+---
+
+## Standard User
+
+Standard users can:
+
+* Manage owned URLs
+* Manage owned landing pages
+* View owned analytics
+* Generate API tokens
+* Manage owned content
+
+Standard users cannot:
+
+* Access other users' content
+* Access administrative endpoints
+* Access system settings
+
+---
+
+## System Accounts
+
+System accounts are internal accounts.
+
+They cannot authenticate into:
+
+* Dashboard
+* REST API
+
+---
+
+# Multi-User Isolation
+
+Multi-user isolation is one of the primary security features of BZOD.
+
+Each tenant receives independent databases.
+
+Example:
+
+```text
+users/
+├── 2/
+│   ├── content.db
+│   └── analytics.db
+│
+├── 3/
+│   ├── content.db
+│   └── analytics.db
+```
+
+User 2 never accesses:
+
+```text
+users/3/content.db
+users/3/analytics.db
+```
+
+User 3 never accesses:
+
+```text
+users/2/content.db
+users/2/analytics.db
+```
+
+---
+
+# Global Slug Security
+
+All public slugs are stored in:
+
+```text
+system.db.global_slugs
+```
+
+Each slug is globally unique.
+
+Example:
+
+```text
+/company
+```
+
+may belong to only one owner.
+
+Duplicate registrations are rejected.
+
+---
+
+## Slug Ownership
+
+Every slug contains:
+
+```text
+owner_user_id
+target_id
+target_type
+status
+```
+
+Ownership must match before modification is permitted.
+
+---
+
+## Slug Transfer Protection
+
+Only administrators may transfer ownership.
+
+Transfer operations:
+
+1. Validate destination quotas
+2. Validate destination user
+3. Copy content
+4. Update ownership
+5. Record history
+6. Write audit event
+
+---
+
+# API Security
+
+REST API authentication uses API tokens.
+
+Tokens are stored as hashes.
+
+Plaintext tokens are shown only once during creation.
+
+---
+
+## API Token Security
+
+Stored values:
+
+```text
+token_hash
+```
+
+Never:
+
+```text
+plaintext_token
+```
+
+If a token is lost:
+
+1. Revoke it
+2. Generate a new token
+
+---
+
+## API Permissions
+
+Admin tokens:
+
+```text
+Full administrative access
+```
+
+Standard user tokens:
+
+```text
+Owned resources only
+```
+
+System accounts:
+
+```text
+API access denied
+```
+
+---
+
+# CSRF Protection
+
+All dashboard forms require valid CSRF tokens.
+
+Protected actions include:
+
+* Login
+* User creation
+* Password reset
+* Content modification
+* Moderation actions
+* Quota updates
+* Backup operations
+
+---
+
+## Invalid CSRF Requests
+
+Invalid requests return:
+
+```http
+403 Forbidden
+```
+
+and are rejected before processing.
+
+---
+
+# XSS Protection
+
+User-supplied content is validated before rendering.
+
+Templates use:
+
+```text
+Askama
+```
+
+which escapes output by default.
+
+Recommended:
+
+* Do not allow arbitrary JavaScript
+* Validate HTML content
+* Restrict trusted editors
+
+---
+
+# Content Moderation
+
+Administrators may:
+
+* Flag content
+* Disable content
+* Delete content
+
+Disabled content returns:
+
+```http
+410 Gone
+```
+
+for:
+
+```text
+/{slug}
+/p/{slug}
+/api/qr/{slug}.png
+/api/qr/{slug}.svg
+```
+
+---
+
+# Redirect Security
+
+The redirect handler validates destination URLs before constructing HTTP Location headers.
+
+Protections include:
+
+* URL scheme validation (only http and https destinations are permitted)
+* Control character rejection
+* CRLF injection prevention
+* Safe Location header construction (no panics on malformed values)
+
+Invalid redirect destinations return:
+
+```http
+500 Internal Server Error
+```
+
+with structured server-side logging. Full destination values are not exposed to clients.
+
+---
+
+# Audit Logging
+
+Security-sensitive actions are logged.
+
+Examples:
+
+```text
+login
+logout
+failed_login
+user_created
+user_deleted
+password_reset
+slug_transfer
+quota_update
+backup_created
+restore_executed
+```
+
+Stored in:
+
+```text
+system.db
+```
+
+Audit logs should be reviewed regularly.
+
+---
+
+# Backup Security
+
+Backups may contain:
+
+* User records
+* Session records
+* URLs
+* Pages
+* Analytics
+* API token hashes
+
+Backups should be treated as sensitive data.
+
+---
+
+## Recommendations
+
+Store backups:
+
+* Offsite
+* Encrypted
+* Access-controlled
+
+Never expose backup archives publicly.
+
+---
+
+# Database Security
+
+SQLite databases should be accessible only to the BZOD service account.
+
+Recommended permissions:
+
+```bash
+chmod 700 data
+chmod 600 *.db
+```
+
+---
+
+# HTTPS Requirements
+
+Production deployments should always use HTTPS.
+
+Recommended reverse proxies:
+
+* Nginx
+* Caddy
+* Traefik
+
+Never expose login pages over plaintext HTTP.
+
+---
+
+# Security Headers
+
+Recommended reverse proxy headers:
+
+```http
+X-Frame-Options: DENY
+X-Content-Type-Options: nosniff
+Referrer-Policy: strict-origin-when-cross-origin
+Content-Security-Policy: default-src 'self'
+```
+
+---
+
+# Password Policy Recommendations
+
+Recommended minimum:
+
+```text
+12 characters
+```
+
+Encourage:
+
+* Password managers
+* Unique passwords
+* Randomly generated credentials
+
+Avoid:
+
+* Reused passwords
+* Dictionary words
+* Predictable patterns
+
+---
+
+# Brute Force Protection
+
+Recommended deployment protections:
+
+* Reverse proxy rate limiting
+* Fail2Ban
+* Firewall rules
+
+Example:
+
+```text
+5 login attempts
+within 5 minutes
+```
+
+before temporary blocking.
+
+---
+
+# Administrative Security Checklist
+
+Before production deployment:
+
+* Enable HTTPS
+* Configure backups
+* Review file permissions
+* Remove default credentials
+* Verify audit logging
+* Test restore procedures
+* Review active sessions
+
+---
+
+# Incident Response
+
+If compromise is suspected:
+
+1. Disable affected accounts.
+2. Revoke active sessions.
+3. Revoke API tokens.
+4. Create forensic backup.
+5. Review audit logs.
+6. Restore from trusted backups if necessary.
+7. Rotate credentials.
+
+---
+
+# Security Testing
+
+BZOD v0.6.0 includes tests covering:
+
+* Authentication
+* Authorization
+* Session validation
+* CSRF enforcement
+* Slug ownership
+* User isolation
+* Upgrade migrations
+* Backup integrity
+* Disaster recovery
+* Redirect destination validation
+* HTTP Location header safety
+
+These tests are executed during CI and release validation.
+
+---
+
+# Responsible Disclosure
+
+If a security vulnerability is discovered:
+
+1. Do not publish exploit details immediately.
+2. Report the issue privately.
+3. Allow time for remediation.
+4. Coordinate disclosure after a fix is available.
+
+---
+
+# Known Limitations
+
+Current limitations include:
+
+* No MFA support
+* No SSO integration
+* No hardware security key support
+* No built-in rate limiter
+* No WebAuthn support
+
+These may be addressed in future releases.
+
+---
+
+# Summary
+
+BZOD v0.6.0 provides:
+
+* Centralized authentication
+* Secure session management
+* RBAC authorization
+* Multi-user isolation
+* Global slug ownership controls
+* CSRF protection
+* API token hashing
+* Audit logging
+* Backup security
+* Operational security guidance
+
+while maintaining a lightweight, SQLite-native, self-hosted architecture.
+
+---
+
+End of Document.
@@ -0,0 +1,516 @@
+# BZOD Testing & Validation Guide
+
+## Overview
+
+BZOD follows a defense-in-depth validation strategy.
+
+A release is considered valid only when:
+
+* Code quality checks pass
+* Automated tests pass
+* Upgrade validation passes
+* Backup/restore validation passes
+* Namespace integrity validation passes
+* Multi-user isolation validation passes
+* Disaster recovery validation passes
+
+The objective is not simply to ensure the application starts, but to ensure that it can be safely upgraded, operated, backed up, restored, and recovered.
+
+---
+
+# Validation Philosophy
+
+BZOD prioritizes:
+
+1. Namespace Integrity
+2. Data Integrity
+3. Multi-Tenant Isolation
+4. Operational Simplicity
+5. Recovery Capability
+6. Security
+7. Functional Correctness
+
+A successful release is not merely one that runs.
+
+A successful release is one that can be recovered.
+
+---
+
+# Automated Test Coverage
+
+Current validation suite includes:
+
+* Unit Tests
+* Integration Tests
+* HTTP E2E Tests
+* Business Workflow Tests
+* Security Tests
+* Backup & Restore Tests
+* Disaster Recovery Tests
+* Migration Tests
+* Upgrade Validation Tests
+* Namespace Integrity Tests
+* Ownership Isolation Tests
+* Dashboard Parity Tests
+* QR Endpoint Tests
+* Concurrency Tests
+* WAL Recovery Tests
+
+The platform currently executes approximately 100+ automated tests.
+
+---
+
+# 1. Build Validation
+
+Verify successful compilation.
+
+```bash
+cargo check
+cargo build
+cargo build --release
+```
+
+Expected:
+
+* No compilation failures
+* Release binary generated
+
+---
+
+# 2. Formatting Validation
+
+```bash
+cargo fmt --check
+```
+
+Expected:
+
+* No formatting errors
+
+---
+
+# 3. Static Analysis
+
+```bash
+cargo clippy --all-targets -- -D warnings
+```
+
+Expected:
+
+* Zero warnings
+* Zero errors
+
+---
+
+# 4. Complete Automated Test Suite
+
+```bash
+cargo test --all-targets -- --nocapture
+```
+
+Expected:
+
+* All tests pass
+* No failures
+* No ignored critical tests
+
+---
+
+# 5. Database Initialization Validation
+
+Create clean environment:
+
+```bash
+rm -rf data
+```
+
+Run:
+
+```bash
+bzod stats
+```
+
+Expected:
+
+* Database hierarchy created
+* Migrations applied
+* System healthy
+
+Validate:
+
+```bash
+bzod doctor
+```
+
+Expected:
+
+```text
+Overall status: HEALTHY
+```
+
+---
+
+# 6. Namespace Integrity Validation
+
+BZOD maintains a global slug namespace.
+
+The following must never coexist:
+
+```text
+Admin URL
+hello
+
+User URL
+hello
+
+Landing Page
+hello
+```
+
+Validate:
+
+```bash
+bzod doctor
+```
+
+Expected:
+
+```text
+No namespace conflicts detected
+```
+
+Duplicate slugs must abort upgrade and restore operations.
+
+---
+
+# 7. Multi-User Isolation Validation
+
+Verify:
+
+* User A cannot access User B URLs
+* User A cannot access User B Pages
+* User A cannot access User B Analytics
+* User A cannot export User B analytics
+
+Expected:
+
+```http
+403 Forbidden
+```
+
+for all unauthorized access.
+
+---
+
+# 8. Dashboard Parity Validation
+
+Verify:
+
+## Administrator URLs
+
+Contains:
+
+* Analytics
+* QR Preview
+* PNG Download
+* SVG Download
+
+## User URLs
+
+Contains identical functionality.
+
+Differences allowed:
+
+* User Management
+* Moderation
+* Backups
+* Health
+* Audit
+* Quotas
+
+Everything else must match.
+
+---
+
+# 9. Analytics Validation
+
+Verify:
+
+* URL Analytics
+* Landing Page Analytics
+* CSV Export
+* JSON Export
+* Date Filters
+* Charts
+* Referrer Breakdown
+* Country Breakdown
+* Browser Breakdown
+* Device Breakdown
+
+Expected:
+
+Administrator and owner views return identical analytics.
+
+---
+
+# 10. QR Validation
+
+Verify:
+
+```text
+/api/qr/<slug>.png
+/api/qr/<slug>.svg
+```
+
+Expected:
+
+```http
+200 OK
+```
+
+Verify:
+
+```text
+Content-Type: image/png
+Content-Type: image/svg+xml
+```
+
+Disabled resources:
+
+```http
+410 Gone
+```
+
+Missing resources:
+
+```http
+404 Not Found
+```
+
+---
+
+# 11. Routing Validation
+
+URL resources:
+
+```text
+/<slug>
+```
+
+must redirect correctly.
+
+Landing Pages:
+
+```text
+/<slug>
+```
+
+must redirect permanently to:
+
+```text
+/p/<slug>
+```
+
+Expected:
+
+```http
+301 Moved Permanently
+```
+
+and:
+
+```http
+200 OK
+```
+
+for final landing page render.
+
+Root landing page:
+
+```text
+GET /
+```
+
+must serve the static landing page.
+
+Expected:
+
+```http
+200 OK
+Content-Type: text/html
+```
+
+Redirect security:
+
+Redirect destinations are validated against:
+
+* Invalid URL schemes
+* CRLF injection attempts
+* Control character injection
+* Malformed HTTP Location header values
+
+Invalid destinations must return:
+
+```http
+500 Internal Server Error
+```
+
+and must not panic or produce malformed HTTP responses.
+
+---
+
+# 12. Backup Validation
+
+Create backup:
+
+```bash
+bzod backup
+```
+
+Expected:
+
+Archive generated successfully.
+
+Validate archive contents.
+
+---
+
+# 13. Restore Validation
+
+Restore backup:
+
+```bash
+bzod restore --file backup.tar.gz
+```
+
+Expected:
+
+* Restore succeeds
+* All data preserved
+* Namespace integrity preserved
+
+---
+
+# 14. Collision Protection Validation
+
+Attempt restore containing duplicate slugs.
+
+Expected:
+
+```text
+Restore aborted
+Slug conflict detected
+```
+
+No partial restore.
+
+---
+
+# 15. Upgrade Validation
+
+Verify upgrade from legacy deployments.
+
+Expected:
+
+* User databases migrated
+* Analytics preserved
+* Links preserved
+* Landing pages preserved
+* Authentication preserved
+
+Duplicate slugs must abort upgrade.
+
+---
+
+# 16. Disaster Recovery Validation
+
+Procedure:
+
+1. Backup system
+2. Stop service
+3. Remove data directory
+4. Restore backup
+5. Start service
+
+Expected:
+
+* Full recovery
+* No manual repair
+* All URLs functional
+* All Landing Pages functional
+* Analytics preserved
+
+---
+
+# 17. Docker Validation
+
+```bash
+docker compose build --no-cache
+docker compose up -d
+```
+
+Verify:
+
+```bash
+docker compose logs
+```
+
+Expected:
+
+```text
+Server started successfully
+```
+
+Container health:
+
+```text
+healthy
+```
+
+---
+
+# 18. WAL Recovery Validation
+
+Verify:
+
+* SQLite WAL mode enabled
+* Recovery after backup succeeds
+* No corruption detected
+
+---
+
+# Release Validation Checklist
+
+Before every release:
+
+```bash
+cargo fmt --check
+
+cargo clippy --all-targets -- -D warnings
+
+cargo test --all-targets -- --nocapture
+
+cargo build --release
+
+cargo audit
+```
+
+Release is approved only if all steps succeed.
+
+---
+
+# Release Blockers
+
+The following are release blockers:
+
+* Namespace conflicts
+* Backup failure
+* Restore failure
+* Upgrade failure
+* Multi-user isolation failure
+* Ownership validation failure
+* Security test failure
+* Data corruption
+* Disaster recovery failure
+
+A release that cannot be restored is not considered production ready.
@@ -0,0 +1,795 @@
+# Upgrade Guide
+
+Version: v0.6.0
+
+This document describes the upgrade process for existing BZOD deployments upgrading to BZOD v0.5.1.
+
+---
+
+# Overview
+
+BZOD v0.5.1 is a platform hardening release focused on:
+
+* Global namespace integrity
+* Multi-tenant safety
+* Dashboard parity
+* QR reliability
+* Upgrade validation
+* Restore collision protection
+* Ownership isolation
+
+While v0.5.0 introduced the multi-user architecture, v0.5.1 strengthens the operational and data integrity guarantees required for production deployments.
+
+---
+
+# Supported Upgrade Paths
+
+Supported:
+
+```text
+v0.5.0 → v0.5.1
+v0.4.x → v0.5.1
+```
+
+Recommended:
+
+```text
+v0.4.x → v0.5.0 → v0.5.1
+```
+
+Unsupported:
+
+```text
+v0.3.x → v0.5.1
+```
+
+Older installations should first upgrade to v0.4.x.
+
+---
+
+# Major Changes in v0.5.1
+
+## Global Namespace Enforcement
+
+BZOD now enforces a single platform-wide slug namespace.
+
+The following resources can no longer share the same slug:
+
+* Administrator URLs
+* Administrator Landing Pages
+* User URLs
+* User Landing Pages
+
+Example:
+
+```text
+Admin URL:
+hello
+
+User URL:
+hello
+```
+
+Result:
+
+```text
+Upgrade aborted.
+Namespace conflict detected.
+```
+
+---
+
+## Global Slug Registry
+
+BZOD now treats the slug registry as the authoritative source of truth.
+
+All slugs are registered in:
+
+```text
+system.db
+```
+
+Table:
+
+```text
+global_slugs
+```
+
+The registry tracks:
+
+```text
+slug
+owner_user_id
+target_type
+target_id
+status
+```
+
+---
+
+## Reservation-Based Slug Allocation
+
+Slug creation now follows:
+
+```text
+Quota Validation
+↓
+Reserve Global Slug
+↓
+Create Resource
+↓
+Activate Slug
+↓
+Update Quotas
+↓
+Audit Log
+```
+
+Benefits:
+
+* Prevents race conditions
+* Prevents duplicate allocations
+* Improves rollback safety
+* Improves multi-user integrity
+
+---
+
+## Stale Reservation Recovery
+
+BZOD automatically cleans abandoned reservations created by:
+
+* Server crashes
+* Interrupted requests
+* Failed transactions
+
+Stale reservations are validated and cleaned during startup.
+
+---
+
+# Breaking Changes
+
+## Global Slug Uniqueness
+
+Deployments containing duplicate slugs will not upgrade.
+
+Example:
+
+```text
+User 1:
+!nx9-dns-server
+
+User 3:
+!nx9-dns-server
+```
+
+Result:
+
+```text
+Upgrade aborted.
+
+Database upgrade aborted due to slug conflicts.
+```
+
+Conflicts must be resolved before migration can continue.
+
+---
+
+## Restore Collision Protection
+
+Restore operations now validate namespace integrity.
+
+Example:
+
+```text
+Existing slug:
+company
+
+Backup slug:
+company
+```
+
+Result:
+
+```text
+Restore aborted.
+Slug conflict detected.
+```
+
+No partial restore occurs.
+
+---
+
+# Pre-Upgrade Checklist
+
+Before upgrading:
+
+* Create backup
+* Verify backup integrity
+* Stop active traffic
+* Run diagnostics
+* Resolve namespace conflicts
+
+---
+
+# Step 1: Create Backup
+
+Full backup:
+
+```bash
+bzod backup
+```
+
+Manual backup:
+
+```bash
+tar czf bzod-backup.tar.gz data/
+```
+
+---
+
+# Step 2: Verify Backup
+
+Verify archive contents:
+
+```text
+users.db
+system.db
+
+users/
+```
+
+If upgrading from legacy versions:
+
+```text
+admin.db
+content.db
+analytics.db
+```
+
+should also be present.
+
+---
+
+# Step 3: Run Diagnostics
+
+Execute:
+
+```bash
+bzod doctor
+```
+
+Expected:
+
+```text
+Overall Status: HEALTHY
+```
+
+Verify:
+
+```text
+No namespace conflicts detected
+No ownership violations detected
+No registry corruption detected
+```
+
+---
+
+# Step 4: Stop Service
+
+Systemd:
+
+```bash
+sudo systemctl stop bzod
+```
+
+Docker:
+
+```bash
+docker compose down
+```
+
+---
+
+# Upgrade Procedure
+
+## Install New Version
+
+Build:
+
+```bash
+cargo build --release
+```
+
+Or install official release binary.
+
+---
+
+## Start BZOD
+
+```bash
+bzod serve
+```
+
+or:
+
+```bash
+docker compose up -d
+```
+
+---
+
+# Automatic Upgrade Actions
+
+During startup BZOD automatically performs:
+
+1. Database migration checks
+2. Namespace integrity validation
+3. Registry validation
+4. Stale reservation cleanup
+5. Global slug verification
+6. Schema migration execution
+
+---
+
+# Namespace Validation
+
+BZOD scans:
+
+```text
+legacy databases
+administrator databases
+tenant databases
+```
+
+for duplicate slugs.
+
+Example:
+
+```text
+Owner 1:
+hello
+
+Owner 3:
+hello
+```
+
+Result:
+
+```text
+Namespace conflict detected.
+Upgrade aborted.
+```
+
+---
+
+# Registry Validation
+
+BZOD validates:
+
+* Duplicate slug entries
+* Missing owners
+* Missing targets
+* Invalid target types
+* Invalid status values
+
+Allowed target types:
+
+```text
+url
+page
+```
+
+Allowed statuses:
+
+```text
+reserving
+active
+disabled
+```
+
+---
+
+# Post-Upgrade Validation
+
+Run:
+
+```bash
+bzod doctor
+```
+
+Expected:
+
+```text
+Namespace Integrity: PASS
+Registry Integrity: PASS
+Ownership Integrity: PASS
+Database Integrity: PASS
+```
+
+---
+
+# Login Validation
+
+Verify:
+
+```text
+Administrator login succeeds
+User login succeeds
+```
+
+---
+
+# URL Validation
+
+Verify:
+
+```text
+https://example.com/abc123
+```
+
+redirects correctly.
+
+Expected:
+
+```http
+302 Found
+```
+
+or configured redirect behavior.
+
+---
+
+# Landing Page Validation
+
+Verify:
+
+```text
+https://example.com/p/demo
+```
+
+renders successfully.
+
+Verify:
+
+```text
+https://example.com/demo
+```
+
+redirects permanently:
+
+```http
+301 Moved Permanently
+```
+
+to:
+
+```text
+/p/demo
+```
+
+---
+
+# QR Validation
+
+Verify:
+
+```text
+/api/qr/demo.png
+/api/qr/demo.svg
+```
+
+Expected:
+
+```http
+200 OK
+```
+
+Content types:
+
+```text
+image/png
+image/svg+xml
+```
+
+Disabled resources:
+
+```http
+410 Gone
+```
+
+Missing resources:
+
+```http
+404 Not Found
+```
+
+---
+
+# Dashboard Validation
+
+Verify Administrator Dashboards:
+
+* URLs
+* Landing Pages
+* Analytics
+* QR Preview
+* PNG Download
+* SVG Download
+
+Verify Standard User Dashboards:
+
+* URLs
+* Landing Pages
+* Analytics
+* QR Preview
+* PNG Download
+* SVG Download
+
+Both should provide equivalent functionality except for administrator-only operations.
+
+---
+
+# Ownership Isolation Validation
+
+Verify:
+
+```text
+User A
+```
+
+cannot access:
+
+```text
+User B Analytics
+User B URLs
+User B Landing Pages
+User B Exports
+```
+
+Expected:
+
+```http
+403 Forbidden
+```
+
+---
+
+# Backup & Restore Validation
+
+Create backup:
+
+```bash
+bzod backup
+```
+
+Restore backup:
+
+```bash
+bzod restore backup.tar.gz
+```
+
+Expected:
+
+* No namespace conflicts
+* No ownership conflicts
+* No partial restores
+
+---
+
+# Rollback Procedure
+
+If upgrade validation fails:
+
+Stop service:
+
+```bash
+sudo systemctl stop bzod
+```
+
+or:
+
+```bash
+docker compose down
+```
+
+Restore backup:
+
+```bash
+bzod restore backup.tar.gz
+```
+
+or restore archived data directory.
+
+Reinstall previous release.
+
+---
+
+# Docker Upgrade
+
+Pull image:
+
+```bash
+docker compose pull
+```
+
+Restart:
+
+```bash
+docker compose up -d
+```
+
+Monitor:
+
+```bash
+docker compose logs -f
+```
+
+Expected:
+
+```text
+Namespace validation passed
+Registry validation passed
+Server started successfully
+```
+
+---
+
+# Systemd Upgrade
+
+Replace binary:
+
+```bash
+sudo cp bzod /usr/local/bin/
+```
+
+Restart:
+
+```bash
+sudo systemctl restart bzod
+```
+
+Verify:
+
+```bash
+sudo systemctl status bzod
+```
+
+Expected:
+
+```text
+active (running)
+```
+
+---
+
+# Automated Upgrade Validation
+
+Execute:
+
+```bash
+cargo fmt --check
+cargo clippy --all-targets -- -D warnings
+cargo test --all-targets -- --nocapture
+```
+
+Particularly validate:
+
+```text
+upgrade_validation_tests
+backup_restore_tests
+slug_registry_tests
+ownership_tests
+analytics_parity_tests
+transaction_tests
+```
+
+---
+
+# Recommended Upgrade Workflow
+
+```text
+1. Create Backup
+2. Verify Backup
+3. Run bzod doctor
+4. Resolve Namespace Conflicts
+5. Stop Service
+6. Install v0.5.1
+7. Start Service
+8. Validate Registry
+9. Validate URLs
+10. Validate Landing Pages
+11. Validate QR Endpoints
+12. Validate Dashboards
+13. Validate Ownership Isolation
+14. Return To Production
+```
+
+---
+
+# Troubleshooting
+
+## Upgrade Aborted Due To Slug Conflicts
+
+Example:
+
+```text
+Slug '!nx9-dns-server'
+is defined in multiple content databases
+by owners [1,3]
+```
+
+Cause:
+
+```text
+Duplicate slug detected.
+```
+
+Resolution:
+
+```text
+Rename or remove conflicting resources.
+Restart upgrade.
+```
+
+---
+
+## QR Codes Return 404
+
+Verify:
+
+```text
+global_slugs
+```
+
+contains the slug.
+
+Verify slug status:
+
+```text
+active
+```
+
+---
+
+## Landing Page Redirect Fails
+
+Verify:
+
+```text
+target_type = page
+```
+
+in:
+
+```text
+global_slugs
+```
+
+---
+
+## Ownership Errors
+
+Run:
+
+```bash
+bzod doctor
+```
+
+Verify ownership integrity passes.
+
+---
+
+# Upgrade Status
+
+BZOD v0.5.1 upgrade path has been validated through:
+
+* Migration Tests
+* Upgrade Validation Tests
+* Namespace Integrity Tests
+* Ownership Isolation Tests
+* Backup & Restore Tests
+* Dashboard Parity Tests
+* QR Endpoint Tests
+* Routing Tests
+
+The v0.5.1 upgrade path is considered production-ready.
@@ -8,15 +8,19 @@ pub struct AnalyticsQueue {
 }

 impl AnalyticsQueue {
-    pub fn new(db: Db, capacity: usize) -> Self {
+    pub fn new(
+        db: Db,
+        capacity: usize,
+        shutdown_rx: tokio::sync::watch::Receiver<bool>,
+    ) -> (Self, tokio::task::JoinHandle<()>) {
        let (sender, receiver) = mpsc::channel(capacity);

        // Spawn background worker to batch-write records
-        tokio::spawn(async move {
-            super::worker::run_worker(db, receiver).await;
+        let handle = tokio::spawn(async move {
+            super::worker::run_worker(db, receiver, shutdown_rx).await;
        });

-        Self { sender }
+        (Self { sender }, handle)
    }

    // Attempt to queue a visit. Non-blocking.
@@ -7,7 +7,11 @@ use crate::db::analytics::insert_visits_batch;
 use crate::db::Db;
 use crate::models::VisitRecord;

-pub async fn run_worker(db: Db, mut receiver: mpsc::Receiver<VisitRecord>) {
+pub async fn run_worker(
+    db: Db,
+    mut receiver: mpsc::Receiver<VisitRecord>,
+    mut shutdown_rx: tokio::sync::watch::Receiver<bool>,
+) {
    let mut batch = Vec::new();
    let batch_size = 50;
    let flush_interval = Duration::from_secs(2);
@@ -37,6 +41,11 @@ pub async fn run_worker(db: Db, mut receiver: mpsc::Receiver<VisitRecord>) {
                    flush_batch(&db, &mut batch);
                }
            }
+            _ = shutdown_rx.changed() => {
+                info!("Analytics worker flushing pending records");
+                flush_batch(&db, &mut batch);
+                break;
+            }
        }
    }
 }
@@ -46,11 +55,47 @@ fn flush_batch(db: &Db, batch: &mut Vec<VisitRecord>) {
        return;
    }

-    info!("Flushing {} visits to analytics database", batch.len());
-    let mut conn_lock = db.analytics.lock().unwrap();
-    if let Err(e) = insert_visits_batch(&mut conn_lock, batch) {
-        error!("Failed to write analytics batch to database: {:?}", e);
-    } else {
-        batch.clear();
+    info!(
+        "Flushing {} visits to user analytics databases",
+        batch.len()
+    );
+
+    // Group visits by owner_user_id
+    let mut groups: std::collections::HashMap<i64, Vec<VisitRecord>> =
+        std::collections::HashMap::new();
+    for record in batch.drain(..) {
+        let user_id = record.owner_user_id.unwrap_or(1); // fallback to legacy_admin (user 1)
+        groups.entry(user_id).or_default().push(record);
+    }
+
+    for (user_id, user_visits) in groups {
+        let db_path = db
+            .data_dir
+            .join("users")
+            .join(user_id.to_string())
+            .join("analytics.db");
+        if let Some(parent) = db_path.parent() {
+            let _ = std::fs::create_dir_all(parent);
+        }
+
+        match rusqlite::Connection::open(&db_path) {
+            Ok(mut conn) => {
+                let _ = crate::db::sqlite::enable_wal(&conn, "analytics");
+                let _ = crate::db::sqlite::enable_foreign_keys(&conn, "analytics");
+
+                if let Err(e) = insert_visits_batch(&mut conn, &user_visits) {
+                    error!(
+                        "Failed to write analytics batch to user {} database: {:?}",
+                        user_id, e
+                    );
+                }
+            }
+            Err(e) => {
+                error!(
+                    "Failed to open analytics database for user {}: {:?}",
+                    user_id, e
+                );
+            }
+        }
    }
 }
@@ -1,13 +1,44 @@
 use crate::auth::session::authenticate_api_key;
-use crate::models::User;
+use crate::models::ApiActor;
 use crate::state::AppState;
 use axum::{
    extract::{FromRef, FromRequestParts},
    http::{request::Parts, StatusCode},
+    Json,
 };

 // Extractor: Authenticate API requests using Bearer token
-pub struct ApiUser(pub User);
+pub struct ApiUser(pub ApiActor);
+
+impl ApiUser {
+    pub fn require_admin(
+        &self,
+    ) -> Result<&crate::models::User, (StatusCode, Json<crate::web::api::ApiError>)> {
+        match &self.0 {
+            ApiActor::Admin(u) => Ok(u),
+            _ => Err((
+                StatusCode::FORBIDDEN,
+                Json(crate::web::api::ApiError {
+                    error: "Admin privileges required".to_string(),
+                }),
+            )),
+        }
+    }
+
+    pub fn require_tenant(
+        &self,
+    ) -> Result<&crate::models::TenantUser, (StatusCode, Json<crate::web::api::ApiError>)> {
+        match &self.0 {
+            ApiActor::User(u) => Ok(u),
+            _ => Err((
+                StatusCode::FORBIDDEN,
+                Json(crate::web::api::ApiError {
+                    error: "Tenant privileges required".to_string(),
+                }),
+            )),
+        }
+    }
+}

 #[axum::async_trait]
 impl<S> FromRequestParts<S> for ApiUser
@@ -25,9 +56,10 @@ where
            .and_then(|h| h.to_str().ok())
            .ok_or((StatusCode::UNAUTHORIZED, "Missing Authorization header"))?;

-        let conn = app_state.admin_db.lock().unwrap();
-        match authenticate_api_key(&conn, auth_header) {
-            Ok(Some(user)) => Ok(ApiUser(user)),
+        let admin_conn = app_state.admin_db.lock().unwrap();
+        let users_conn = app_state.users_db.lock().unwrap();
+        match authenticate_api_key(&admin_conn, &users_conn, auth_header) {
+            Ok(Some(actor)) => Ok(ApiUser(actor)),
            Ok(None) => Err((StatusCode::UNAUTHORIZED, "Invalid API token")),
            Err(_) => Err((StatusCode::INTERNAL_SERVER_ERROR, "Database error")),
        }
@@ -6,4 +6,6 @@ pub mod session;
 pub use csrf::{generate_csrf_token, verify_csrf};
 pub use middleware::ApiUser;
 pub use password::{hash_password, verify_password, verify_sha256};
-pub use session::{authenticate_api_key, authenticate_session, generate_token};
+pub use session::{
+    authenticate_admin_session, authenticate_api_key, authenticate_user_session, generate_token,
+};
@@ -1,11 +1,11 @@
 use crate::db::admin::{
-    get_api_key_by_hash, get_session, get_user_by_id, update_api_key_last_used,
+    get_api_key_by_hash, get_user_by_id as get_admin_user_by_id, update_api_key_last_used,
 };
-use crate::models::User;
+use crate::models::{ApiActor, Session as AdminSession, TenantUser, User, UserSession};
 use axum_extra::extract::CookieJar;
 use chrono::Utc;
 use rand::{thread_rng, RngCore};
-use rusqlite::Connection;
+use rusqlite::{Connection, OptionalExtension};
 use sha2::{Digest, Sha256};

 // Generate a secure random token (hex-encoded)
@@ -15,8 +15,8 @@ pub fn generate_token(bytes_len: usize) -> String {
    hex::encode(key)
 }

-// Authenticate session from cookies
-pub fn authenticate_session(
+// Authenticate administrator session from cookies
+pub fn authenticate_admin_session(
    conn: &Connection,
    jar: &CookieJar,
 ) -> Result<Option<(User, String)>, rusqlite::Error> {
@@ -26,7 +26,33 @@ pub fn authenticate_session(
    };

    let session_id = cookie.value();
-    let session = match get_session(conn, session_id)? {
+    let session_opt: Option<AdminSession> = conn
+        .query_row(
+            "SELECT id, user_id, expires_at, created_at FROM sessions WHERE id = ?1;",
+            [session_id],
+            |row| {
+                let id: String = row.get(0)?;
+                // `user_id` may be stored as integer (users.db) or text (admin.db UUID).
+                let user_id_str: String = match row.get::<_, String>(1) {
+                    Ok(s) => s,
+                    Err(_) => {
+                        let i: i64 = row.get(1)?;
+                        i.to_string()
+                    }
+                };
+                let expires_at: String = row.get(2)?;
+                let created_at: String = row.get(3)?;
+                Ok(AdminSession {
+                    id,
+                    user_id: user_id_str,
+                    expires_at,
+                    created_at,
+                })
+            },
+        )
+        .optional()?;
+
+    let session = match session_opt {
        Some(s) => s,
        None => return Ok(None),
    };
@@ -41,19 +67,171 @@ pub fn authenticate_session(
        return Ok(None);
    }

-    // Get user
-    if let Some(user) = get_user_by_id(conn, &session.user_id)? {
+    // Get user (status must be 'active' and account_type = 'admin')
+    // If the session user_id looks numeric, bind as integer when querying users.db
+    // Try the extended lookup but catch errors (e.g., missing columns in legacy admin DB)
+    // Try the extended lookup; if it errors (legacy schema), perform a fallback lookup.
+    let (user_opt, extended_failed) = match if let Ok(id_i64) = session.user_id.parse::<i64>() {
+        conn.query_row(
+            "SELECT id, username, password_hash, created_at 
+             FROM users WHERE id = ?1 AND status = 'active' AND account_type = 'admin';",
+            [id_i64],
+            |row| {
+                let id_str = row.get::<_, i64>(0)?.to_string();
+                Ok(User {
+                    id: id_str,
+                    username: row.get(1)?,
+                    password_hash: row.get(2)?,
+                    created_at: row.get(3)?,
+                })
+            },
+        )
+        .optional()
+    } else {
+        conn.query_row(
+            "SELECT id, username, password_hash, created_at 
+             FROM users WHERE id = ?1 AND status = 'active' AND account_type = 'admin';",
+            [session.user_id.as_str()],
+            |row| {
+                // admin DB stores UUID string ids, so read as String
+                let id_str: String = row.get(0)?;
+                Ok(User {
+                    id: id_str,
+                    username: row.get(1)?,
+                    password_hash: row.get(2)?,
+                    created_at: row.get(3)?,
+                })
+            },
+        )
+        .optional()
+    } {
+        Ok(opt) => (opt, false),
+        Err(_) => (None, true),
+    };
+
+    if let Some(user) = user_opt {
+        return Ok(Some((user, session.id)));
+    }
+
+    if extended_failed {
+        // Fallback for legacy admin.db schemas which may not have `status`/`account_type` columns
+        // Try a simpler lookup by id only.
+        let fallback_user_opt = if let Ok(id_i64) = session.user_id.parse::<i64>() {
+            conn.query_row(
+                "SELECT id, username, password_hash, created_at FROM users WHERE id = ?1;",
+                [id_i64],
+                |row| {
+                    Ok(User {
+                        id: row.get::<_, i64>(0)?.to_string(),
+                        username: row.get(1)?,
+                        password_hash: row.get(2)?,
+                        created_at: row.get(3)?,
+                    })
+                },
+            )
+            .optional()
+            .unwrap_or(None)
+        } else {
+            conn.query_row(
+                "SELECT id, username, password_hash, created_at FROM users WHERE id = ?1;",
+                [session.user_id.as_str()],
+                |row| {
+                    Ok(User {
+                        id: row.get(0)?,
+                        username: row.get(1)?,
+                        password_hash: row.get(2)?,
+                        created_at: row.get(3)?,
+                    })
+                },
+            )
+            .optional()
+            .unwrap_or(None)
+        };
+
+        if let Some(user) = fallback_user_opt {
+            Ok(Some((user, session.id)))
+        } else {
+            Ok(None)
+        }
+    } else {
+        Ok(None)
+    }
+}
+
+// Authenticate user session from cookies
+pub fn authenticate_user_session(
+    users_conn: &Connection,
+    jar: &CookieJar,
+) -> Result<Option<(TenantUser, String)>, rusqlite::Error> {
+    let cookie = match jar.get("bzod_user_session") {
+        Some(c) => c,
+        None => return Ok(None),
+    };
+
+    let session_id = cookie.value();
+
+    // Get session from sessions table in users.db
+    let mut stmt = users_conn
+        .prepare("SELECT id, user_id, expires_at, created_at FROM sessions WHERE id = ?1;")?;
+    let session_opt: Option<UserSession> = stmt
+        .query_row([session_id], |row| {
+            Ok(UserSession {
+                id: row.get(0)?,
+                user_id: row.get(1)?,
+                expires_at: row.get(2)?,
+                created_at: row.get(3)?,
+            })
+        })
+        .optional()?;
+
+    let session = match session_opt {
+        Some(s) => s,
+        None => return Ok(None),
+    };
+
+    // Check expiration
+    if let Ok(expires) = chrono::DateTime::parse_from_rfc3339(&session.expires_at) {
+        if expires.with_timezone(&Utc) < Utc::now() {
+            return Ok(None);
+        }
+    } else {
+        return Ok(None);
+    }
+
+    // Get tenant user (status must be 'active')
+    let mut stmt = users_conn.prepare(
+        "SELECT id, username, password_hash, status, created_at, last_login, account_type, organization_id, metadata 
+         FROM users WHERE id = ?1 AND status = 'active';"
+    )?;
+    let user_opt = stmt
+        .query_row([session.user_id], |row| {
+            Ok(TenantUser {
+                id: row.get(0)?,
+                username: row.get(1)?,
+                password_hash: row.get(2)?,
+                status: row.get(3)?,
+                created_at: row.get(4)?,
+                last_login: row.get(5)?,
+                account_type: row.get(6)?,
+                organization_id: row.get(7)?,
+                metadata: row.get(8)?,
+            })
+        })
+        .optional()?;
+
+    if let Some(user) = user_opt {
        Ok(Some((user, session.id)))
    } else {
        Ok(None)
    }
 }

-// Authenticate API key from Authorization header
+// Authenticate API key/token from Authorization header (unified)
 pub fn authenticate_api_key(
-    conn: &Connection,
+    admin_conn: &Connection,
+    users_conn: &Connection,
    auth_header: &str,
-) -> Result<Option<User>, rusqlite::Error> {
+) -> Result<Option<ApiActor>, rusqlite::Error> {
    if !auth_header.starts_with("Bearer ") {
        return Ok(None);
    }
@@ -68,13 +246,73 @@ pub fn authenticate_api_key(
    hasher.update(key.as_bytes());
    let hashed_key = hex::encode(hasher.finalize());

-    if let Some(api_key_rec) = get_api_key_by_hash(conn, &hashed_key)? {
-        // Update last used timestamp
-        update_api_key_last_used(conn, &api_key_rec.id)?;
+    // 1. Check user API tokens in users.db
+    let mut stmt = users_conn.prepare("SELECT user_id FROM api_tokens WHERE token_hash = ?1;")?;
+    let user_id_opt: Option<i64> = stmt.query_row([&hashed_key], |row| row.get(0)).optional()?;

-        // Get user
-        if let Some(user) = get_user_by_id(conn, &api_key_rec.user_id)? {
-            return Ok(Some(user));
+    if let Some(user_id) = user_id_opt {
+        let mut stmt = users_conn.prepare(
+            "SELECT id, username, password_hash, status, created_at, last_login, account_type, organization_id, metadata 
+             FROM users WHERE id = ?1 AND status = 'active';"
+        )?;
+        let user_opt = stmt
+            .query_row([user_id], |row| {
+                Ok(TenantUser {
+                    id: row.get(0)?,
+                    username: row.get(1)?,
+                    password_hash: row.get(2)?,
+                    status: row.get(3)?,
+                    created_at: row.get(4)?,
+                    last_login: row.get(5)?,
+                    account_type: row.get(6)?,
+                    organization_id: row.get(7)?,
+                    metadata: row.get(8)?,
+                })
+            })
+            .optional()?;
+
+        if let Some(user) = user_opt {
+            return Ok(Some(ApiActor::User(user)));
+        }
+    }
+
+    // 2. Check admin system API keys in admin.db
+    if let Some(api_key_rec) = get_api_key_by_hash(admin_conn, &hashed_key)? {
+        // Update last used timestamp
+        update_api_key_last_used(admin_conn, &api_key_rec.id)?;
+
+        // Get admin user from users.db (users_conn)
+        // Try to interpret the api_key user_id as an integer referencing users.db
+        if let Ok(user_id_i64) = api_key_rec.user_id.parse::<i64>() {
+            let mut stmt = users_conn.prepare(
+                "SELECT id, username, password_hash, created_at 
+                 FROM users WHERE id = ?1 AND status = 'active' AND account_type = 'admin';",
+            )?;
+            let user_opt = stmt
+                .query_row([user_id_i64], |row| {
+                    let id_i64: i64 = row.get(0)?;
+                    Ok(User {
+                        id: id_i64.to_string(),
+                        username: row.get(1)?,
+                        password_hash: row.get(2)?,
+                        created_at: row.get(3)?,
+                    })
+                })
+                .optional()?;
+
+            if let Some(user) = user_opt {
+                return Ok(Some(ApiActor::Admin(user)));
+            }
+        }
+
+        // Fallback: admin DB may store users with string UUIDs. Try looking up directly in admin_conn.
+        if let Ok(Some(admin_user)) = get_admin_user_by_id(admin_conn, &api_key_rec.user_id) {
+            return Ok(Some(ApiActor::Admin(User {
+                id: admin_user.id,
+                username: admin_user.username,
+                password_hash: admin_user.password_hash,
+                created_at: admin_user.created_at,
+            })));
        }
    }

@@ -0,0 +1,152 @@
+use crate::config::Config;
+use crate::db::Db;
+use rusqlite::Connection;
+use std::path::PathBuf;
+use tracing::{error, info};
+
+pub async fn run(
+    target_admin_id: i64,
+    data_dir: Option<String>,
+    dry_run: bool,
+    force: bool,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    // 1. Verify target admin exists and is an admin
+    let target_user = {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::get_user_by_id(&conn, target_admin_id)?
+    };
+
+    let target_user = match target_user {
+        Some(u) => u,
+        None => {
+            error!("Target admin ID {} not found", target_admin_id);
+            return Ok(());
+        }
+    };
+
+    if target_user.account_type != "admin" {
+        error!(
+            "Target user '{}' (ID {}) is not an admin account.",
+            target_user.username, target_admin_id
+        );
+        return Ok(());
+    }
+
+    if target_admin_id == 1 {
+        error!("Target admin ID cannot be 1 (legacy admin).");
+        return Ok(());
+    }
+
+    // 2. Open databases
+    let legacy_content_path = config.data_dir.join("users").join("1").join("content.db");
+
+    if !legacy_content_path.exists() {
+        info!(
+            "No legacy admin content database found at {:?}",
+            legacy_content_path
+        );
+        return Ok(());
+    }
+
+    db.init_user_databases(target_admin_id)?;
+    let target_content_path = config
+        .data_dir
+        .join("users")
+        .join(target_admin_id.to_string())
+        .join("content.db");
+
+    let mut legacy_conn = Connection::open(&legacy_content_path)?;
+    let mut target_conn = Connection::open(&target_content_path)?;
+    let mut system_conn = db.system.lock().unwrap();
+
+    println!("Scanning legacy admin content database...");
+
+    // 3. Count items
+    let urls = {
+        let mut stmt = legacy_conn.prepare("SELECT * FROM urls;")?;
+        let mut rows = stmt.query([])?;
+        let mut data = Vec::new();
+        while let Ok(Some(_)) = rows.next() {
+            data.push(1);
+        }
+        data
+    };
+    let url_count = urls.len();
+
+    let pages = {
+        let mut stmt = legacy_conn.prepare("SELECT * FROM landing_pages;")?;
+        let mut rows = stmt.query([])?;
+        let mut data = Vec::new();
+        while let Ok(Some(_)) = rows.next() {
+            data.push(1);
+        }
+        data
+    };
+    let page_count = pages.len();
+
+    println!(
+        "Found {} URLs and {} Landing Pages owned by legacy admin (ID 1).",
+        url_count, page_count
+    );
+
+    if dry_run {
+        println!("Dry run mode enabled. No changes will be made.");
+        return Ok(());
+    }
+
+    if !force {
+        println!("Migration requires the --force flag to execute. Aborting.");
+        return Ok(());
+    }
+
+    println!(
+        "Starting migration to Admin '{}' (ID {})...",
+        target_user.username, target_admin_id
+    );
+
+    // 4. Perform Migration (using ATTACH DATABASE for fast copy)
+    // We attach the legacy db to the target db to do INSERT INTO ... SELECT * FROM
+    target_conn.execute(
+        "ATTACH DATABASE ?1 AS legacy;",
+        rusqlite::params![legacy_content_path.to_string_lossy()],
+    )?;
+
+    let tx = target_conn.transaction()?;
+    tx.execute("INSERT OR IGNORE INTO urls SELECT * FROM legacy.urls;", [])?;
+    tx.execute(
+        "INSERT OR IGNORE INTO landing_pages SELECT * FROM legacy.landing_pages;",
+        [],
+    )?;
+    tx.commit()?;
+
+    target_conn.execute("DETACH DATABASE legacy;", [])?;
+
+    // 5. Update global registry
+    let sys_tx = system_conn.transaction()?;
+    let updated_slugs = sys_tx.execute(
+        "UPDATE global_slugs SET owner_user_id = ?1 WHERE owner_user_id = 1;",
+        rusqlite::params![target_admin_id],
+    )?;
+    sys_tx.commit()?;
+
+    // 6. Delete from legacy
+    let legacy_tx = legacy_conn.transaction()?;
+    legacy_tx.execute("DELETE FROM urls;", [])?;
+    legacy_tx.execute("DELETE FROM landing_pages;", [])?;
+    legacy_tx.commit()?;
+
+    println!("Migration Complete!");
+    println!("-------------------");
+    println!("Migrated {} URLs.", url_count);
+    println!("Migrated {} Landing Pages.", page_count);
+    println!("Updated {} slugs in global registry.", updated_slugs);
+    println!("Cleared legacy content database.");
+
+    Ok(())
+}
@@ -0,0 +1,33 @@
+use crate::config::Config;
+use crate::db::Db;
+use crate::services::destination_audit::{audit_all_destinations, format_report};
+use std::path::PathBuf;
+use tracing::info;
+
+/// Read-only audit of all stored redirect destinations.
+///
+/// Does not rewrite, delete, or "repair" any records.
+pub async fn run(
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+
+    info!("Starting read-only destination audit...");
+    let db = Db::init(&config)?;
+    let report = audit_all_destinations(&db)?;
+    print!("{}", format_report(&report));
+
+    if report.invalid > 0 {
+        // Non-zero exit so automation can detect findings without treating them as crashes.
+        Err(format!(
+            "destination audit found {} invalid stored URL(s)",
+            report.invalid
+        )
+        .into())
+    } else {
+        Ok(())
+    }
+}
@@ -0,0 +1,150 @@
+use crate::config::Config;
+use crate::db::Db;
+use chrono::Utc;
+use std::fs::File;
+use std::path::{Path, PathBuf};
+use tar::{Builder, Header};
+use tracing::{error, info};
+use zstd::Encoder;
+
+#[derive(serde::Serialize, serde::Deserialize)]
+struct UserBackupMetadata {
+    id: i64,
+    username: String,
+    password_hash: String,
+    status: String,
+    created_at: String,
+    account_type: String,
+    metadata: Option<String>,
+    quotas: UserBackupQuotas,
+}
+
+#[derive(serde::Serialize, serde::Deserialize)]
+struct UserBackupQuotas {
+    max_urls: i64,
+    max_landings: i64,
+    max_api_tokens: i64,
+    max_storage_mb: i64,
+}
+
+pub async fn run(
+    username: String,
+    out: Option<String>,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    let username_clean = username.trim().to_lowercase();
+
+    // 1. Get user details from users.db
+    let user_details = {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::get_user_by_username(&conn, &username_clean)?
+    };
+
+    let user = match user_details {
+        Some(u) => u,
+        None => {
+            error!("User '{}' not found", username_clean);
+            return Ok(());
+        }
+    };
+
+    let user_id = user.id;
+
+    // 2. Fetch user's quotas
+    let quotas = {
+        let conn = db.users.lock().unwrap();
+        conn.query_row(
+            "SELECT max_urls, max_landings, max_api_tokens, max_storage_mb FROM quotas WHERE user_id = ?1;",
+            [user_id],
+            |row| {
+                Ok(UserBackupQuotas {
+                    max_urls: row.get(0)?,
+                    max_landings: row.get(1)?,
+                    max_api_tokens: row.get(2)?,
+                    max_storage_mb: row.get(3)?,
+                })
+            }
+        )?
+    };
+
+    // 3. Define output path
+    let tar_path = match out {
+        Some(p) => PathBuf::from(p),
+        None => {
+            if !config.backup_dir.exists() {
+                std::fs::create_dir_all(&config.backup_dir)?;
+            }
+            config.backup_dir.join(format!(
+                "{}-{}.tar.zst",
+                username_clean,
+                Utc::now().format("%Y%m%d")
+            ))
+        }
+    };
+
+    info!(
+        "Backing up user {} (ID: {}) to {:?}",
+        username_clean, user_id, tar_path
+    );
+
+    // 4. Force checkpoint on user's databases
+    let user_dir = config.data_dir.join("users").join(user_id.to_string());
+    if let Ok(c) = rusqlite::Connection::open(user_dir.join("content.db")) {
+        let _ = c.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+    }
+    if let Ok(c) = rusqlite::Connection::open(user_dir.join("analytics.db")) {
+        let _ = c.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+    }
+    if let Ok(c) = rusqlite::Connection::open(user_dir.join("profile.db")) {
+        let _ = c.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+    }
+
+    // 5. Create tar.zst archive
+    let file = File::create(&tar_path)?;
+    let zst_enc = Encoder::new(file, 3)?;
+    let mut tar = Builder::new(zst_enc);
+
+    // Write metadata.json directly into tar
+    let metadata_obj = UserBackupMetadata {
+        id: user.id,
+        username: user.username,
+        password_hash: user.password_hash,
+        status: user.status,
+        created_at: user.created_at,
+        account_type: user.account_type,
+        metadata: user.metadata,
+        quotas,
+    };
+    let metadata_bytes = serde_json::to_vec_pretty(&metadata_obj)?;
+    let mut header = Header::new_gnu();
+    header.set_size(metadata_bytes.len() as u64);
+    header.set_path("metadata.json")?;
+    header.set_mode(0o644);
+    header.set_cksum();
+    tar.append(&header, &metadata_bytes[..])?;
+
+    // Append database files
+    let mut append_file =
+        |name_in_archive: &str, path_on_disk: &Path| -> Result<(), Box<dyn std::error::Error>> {
+            if path_on_disk.exists() {
+                let mut file = File::open(path_on_disk)?;
+                tar.append_file(name_in_archive, &mut file)?;
+            }
+            Ok(())
+        };
+
+    append_file("content.db", &user_dir.join("content.db"))?;
+    append_file("analytics.db", &user_dir.join("analytics.db"))?;
+    append_file("profile.db", &user_dir.join("profile.db"))?;
+
+    tar.into_inner()?.finish()?;
+
+    info!("User backup generated successfully at {:?}", tar_path);
+    Ok(())
+}
@@ -32,8 +32,9 @@ pub async fn run(
    }

    let hash = hash_password(&password).map_err(|e| e.to_string())?;
-    let conn = db.admin.lock().unwrap();
-    let u = crate::db::admin::create_user(&conn, &final_username, &hash)?;
+    let conn = db.users.lock().unwrap();
+    let u = crate::db::users::create_admin_user(&conn, &final_username, &hash)?;
+    db.init_user_databases(u.id)?;
    info!(
        "Successfully created admin user: {} (ID: {})",
        u.username, u.id
@@ -0,0 +1,86 @@
+use crate::auth::hash_password;
+use crate::config::Config;
+use crate::db::Db;
+use std::io::{self, Write};
+use std::path::PathBuf;
+use tracing::{error, info};
+
+pub async fn run(
+    username: Option<String>,
+    password: Option<String>,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    let final_username = match username {
+        Some(u) => u,
+        None => read_input("Enter username: "),
+    };
+
+    let username_clean = final_username.trim().to_lowercase();
+    if username_clean.is_empty() {
+        error!("Username cannot be empty");
+        return Ok(());
+    }
+
+    if username_clean.len() < 3 {
+        error!("Username must be at least 3 characters");
+        return Ok(());
+    }
+
+    if !username_clean
+        .chars()
+        .all(|c| c.is_alphanumeric() || c == '-' || c == '_')
+    {
+        error!("Username must contain only alphanumeric characters, hyphens, or underscores");
+        return Ok(());
+    }
+
+    let final_password = match password {
+        Some(p) => p,
+        None => read_input("Enter password: "),
+    };
+    if final_password.trim().is_empty() {
+        error!("Password cannot be empty");
+        return Ok(());
+    }
+
+    let hash = hash_password(&final_password).map_err(|e| e.to_string())?;
+
+    // Check if user already exists
+    {
+        let conn = db.users.lock().unwrap();
+        if crate::db::users::get_user_by_username(&conn, &username_clean)?.is_some() {
+            error!("User already exists: {}", username_clean);
+            return Ok(());
+        }
+    }
+
+    // Create user in DB (this seeds default quotas too)
+    let new_user = {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::create_user(&conn, &username_clean, &hash, "standard", None)?
+    };
+
+    // Initialize their user specific directory and DB files (content.db, analytics.db, profile.db)
+    db.init_user_databases(new_user.id)?;
+
+    info!(
+        "Successfully created standard user: {} (ID: {})",
+        new_user.username, new_user.id
+    );
+
+    Ok(())
+}
+
+fn read_input(prompt: &str) -> String {
+    print!("{}", prompt);
+    let _ = io::stdout().flush();
+    let mut input = String::new();
+    let _ = io::stdin().read_line(&mut input);
+    input.trim().to_string()
+}
@@ -0,0 +1,92 @@
+use crate::config::Config;
+use crate::db::Db;
+use chrono::Utc;
+use std::path::PathBuf;
+use tracing::{error, info};
+
+pub async fn run(
+    user_id: i64,
+    force: bool,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    if user_id == 1 && !force {
+        error!("Deleting legacy_admin system account requires --force flag");
+        return Ok(());
+    }
+
+    // Capture user details
+    let user_details = {
+        let conn = db.users.lock().unwrap();
+        match crate::db::users::get_user_by_id(&conn, user_id)? {
+            Some(u) => u,
+            None => {
+                error!("User ID {} not found", user_id);
+                return Ok(());
+            }
+        }
+    };
+
+    // 1. Transactional clean up on system.db (deleting their global slug mappings)
+    {
+        let mut system_conn = db.system.lock().unwrap();
+        let tx = system_conn.transaction()?;
+
+        // Get all slugs owned by the user
+        let slugs: Vec<String> = {
+            let mut stmt = tx.prepare("SELECT slug FROM global_slugs WHERE owner_user_id = ?1;")?;
+            let rows = stmt.query_map([user_id], |row| row.get(0))?;
+            rows.filter_map(|r| r.ok()).collect()
+        };
+
+        // Delete from global_slugs and write to history
+        let now = Utc::now().to_rfc3339();
+        for slug in slugs {
+            let _ = tx.execute("DELETE FROM global_slugs WHERE slug = ?1;", [&slug]);
+            let _ = tx.execute(
+                "INSERT INTO slug_history (slug, old_owner_user_id, new_owner_user_id, action, timestamp, admin_username)
+                 VALUES (?1, ?2, NULL, 'deleted', ?3, ?4);",
+                rusqlite::params![slug, user_id, now, "cli"],
+            );
+        }
+
+        tx.commit()?;
+    }
+
+    // 2. Delete user folder and database files from disk
+    let user_dir = config.data_dir.join("users").join(user_id.to_string());
+    if user_dir.exists() {
+        let _ = std::fs::remove_dir_all(&user_dir);
+    }
+
+    // 3. Remove user entry from users.db (cascading deletes quotas/sessions/tokens)
+    {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::delete_user(&conn, user_id)?;
+    }
+
+    // Write audit event
+    {
+        let system_conn = db.system.lock().unwrap();
+        let _ = crate::db::audit_events::write_audit_event(
+            &system_conn,
+            "cli",
+            "USER_DELETION",
+            "user",
+            &user_id.to_string(),
+            Some(&format!("Username: {}", user_details.username)),
+        );
+    }
+
+    info!(
+        "Successfully deleted user {} (ID: {}) and all associated content",
+        user_details.username, user_id
+    );
+
+    Ok(())
+}
@@ -0,0 +1,55 @@
+use crate::config::Config;
+use crate::db::Db;
+use std::path::PathBuf;
+use tracing::{error, info};
+
+pub async fn run(
+    user_id: i64,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    // Check user exists
+    let user = {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::get_user_by_id(&conn, user_id)?
+    };
+
+    let user = match user {
+        Some(u) => u,
+        None => {
+            error!("User ID {} not found", user_id);
+            return Ok(());
+        }
+    };
+
+    if user.status == "disabled" {
+        info!("User {} is already disabled", user.username);
+        return Ok(());
+    }
+
+    {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::update_user_status(&conn, user_id, "disabled")?;
+    }
+
+    // Write audit event
+    {
+        let system_conn = db.system.lock().unwrap();
+        let _ = crate::db::audit_events::write_audit_event(
+            &system_conn,
+            "cli",
+            "USER_DISABLED",
+            "user",
+            &user_id.to_string(),
+            Some(&format!("Username: {}", user.username)),
+        );
+    }
+
+    info!("User {} (ID: {}) has been disabled", user.username, user_id);
+    Ok(())
+}
@@ -22,11 +22,25 @@ pub async fn run(
    println!("Data directory: {:?}", config.data_dir);
    println!();

-    let databases = ["admin", "content", "analytics", "system"];
    let mut all_healthy = true;

-    for db_name in &databases {
-        let db_path = config.data_dir.join(format!("{}.db", db_name));
+    // Define target databases in the new layout
+    let admin_dir = config.data_dir.join("admin");
+    let legacy_user_dir = config.data_dir.join("users").join("1");
+
+    let dbs = vec![
+        ("admin", admin_dir.join("admin.db")),
+        ("system", admin_dir.join("system.db")),
+        ("users", admin_dir.join("users.db")),
+        ("legacy content", legacy_user_dir.join("content.db")),
+        ("legacy analytics", legacy_user_dir.join("analytics.db")),
+    ];
+
+    for (db_name, db_path) in dbs {
+        // Skip legacy databases if they don't exist
+        if db_name.starts_with("legacy") && !db_path.exists() {
+            continue;
+        }

        if !db_path.exists() {
            println!("Database: {}", db_name);
@@ -75,6 +89,95 @@ pub async fn run(
        println!();
    }

+    // Global Slug Registry Integrity Check
+    println!("Global Slug Registry Integrity Check");
+    println!("====================================");
+    let system_db_path = admin_dir.join("system.db");
+    let users_db_path = admin_dir.join("users.db");
+
+    if system_db_path.exists() && users_db_path.exists() {
+        match (
+            Connection::open(&system_db_path),
+            Connection::open(&users_db_path),
+        ) {
+            (Ok(sys_conn), Ok(usr_conn)) => {
+                match crate::services::registry_validator::RegistryValidator::scan(
+                    &sys_conn,
+                    &usr_conn,
+                    &config.data_dir,
+                    None,
+                ) {
+                    Ok(issues) => {
+                        if issues.is_empty() {
+                            println!("  Status: HEALTHY (no issues found)");
+                        } else {
+                            println!("  Status: ISSUES DETECTED");
+                            all_healthy = false;
+
+                            for issue in &issues {
+                                println!();
+                                println!("ERROR");
+                                println!();
+                                println!("Slug:");
+                                println!("    {}", issue.slug);
+                                println!();
+                                println!("Type:");
+                                println!(
+                                    "    {}",
+                                    if issue.target_type == "url" {
+                                        "URL"
+                                    } else if issue.target_type == "page" {
+                                        "Landing Page"
+                                    } else {
+                                        &issue.target_type
+                                    }
+                                );
+                                println!();
+                                println!("Owner:");
+                                println!("    User ID {}", issue.owner_user_id);
+                                println!();
+                                println!("Database:");
+                                println!("    {}", issue.database_path.display());
+                                println!();
+                                println!("Target UUID:");
+                                println!("    {}", issue.target_id);
+                                println!();
+                                println!("Issue:");
+                                println!("    {:?}", issue.issue_type);
+                                println!();
+                                println!("Description:");
+                                println!("    {}", issue.description);
+                                println!();
+                                println!("Suggested Repair:");
+                                println!();
+                                if issue.slug != "*" {
+                                    println!(
+                                        "    bzod repair registry --slug {} --dry-run",
+                                        issue.slug
+                                    );
+                                } else {
+                                    println!("    bzod repair registry --dry-run");
+                                }
+                                println!("--------------------");
+                            }
+                        }
+                    }
+                    Err(e) => {
+                        println!("  Status: ERROR running registry scan: {}", e);
+                        all_healthy = false;
+                    }
+                }
+            }
+            _ => {
+                println!("  Status: ERROR opening system.db or users.db for integrity check");
+                all_healthy = false;
+            }
+        }
+    } else {
+        println!("  Status: SKIPPED (system.db/users.db not found)");
+    }
+    println!();
+
    println!("--------------------");
    if all_healthy {
        println!("Overall status: HEALTHY");
@@ -0,0 +1,58 @@
+use crate::config::Config;
+use crate::db::Db;
+use std::path::PathBuf;
+use tracing::{error, info};
+
+pub async fn run(
+    user_id: i64,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    // Check user exists
+    let user = {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::get_user_by_id(&conn, user_id)?
+    };
+
+    let user = match user {
+        Some(u) => u,
+        None => {
+            error!("User ID {} not found", user_id);
+            return Ok(());
+        }
+    };
+
+    if user.status == "active" {
+        info!("User {} is already active", user.username);
+        return Ok(());
+    }
+
+    {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::update_user_status(&conn, user_id, "active")?;
+    }
+
+    // Write audit event
+    {
+        let system_conn = db.system.lock().unwrap();
+        let _ = crate::db::audit_events::write_audit_event(
+            &system_conn,
+            "cli",
+            "USER_ENABLED",
+            "user",
+            &user_id.to_string(),
+            Some(&format!("Username: {}", user.username)),
+        );
+    }
+
+    info!(
+        "User {} (ID: {}) has been enabled (active)",
+        user.username, user_id
+    );
+    Ok(())
+}
@@ -0,0 +1,32 @@
+use crate::config::Config;
+use crate::db::Db;
+use std::path::PathBuf;
+
+pub async fn run(
+    code: String,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    let normalized_code = code.trim().to_lowercase();
+    if !crate::utils::validation::validate_redirect_code(&normalized_code) {
+        return Err("Invalid short code or custom slug format".into());
+    }
+
+    let url_opt = {
+        let conn = db.content.lock().unwrap();
+        crate::db::content::get_url_by_code(&conn, &normalized_code)?
+    };
+
+    match url_opt {
+        Some(url) => {
+            println!("{}", url.destination);
+            Ok(())
+        }
+        None => Err(format!("Short code not found: {}", normalized_code).into()),
+    }
+}
@@ -0,0 +1,36 @@
+use crate::config::Config;
+use crate::db::Db;
+use std::path::PathBuf;
+
+pub async fn run(
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    let users = {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::list_users(&conn)?
+    };
+
+    println!(
+        "{:<6} | {:<20} | {:<10} | {:<12} | {:<24}",
+        "ID", "Username", "Status", "Type", "Created At"
+    );
+    println!(
+        "{:-<6}-+-{:-<20}-+-{:-<10}-+-{:-<12}-+-{:-<24}",
+        "", "", "", "", ""
+    );
+
+    for u in users {
+        println!(
+            "{:<6} | {:<20} | {:<10} | {:<12} | {:<24}",
+            u.id, u.username, u.status, u.account_type, u.created_at
+        );
+    }
+
+    Ok(())
+}
@@ -1,16 +1,30 @@
 use clap::{Parser, Subcommand};

+pub mod admin_migrate;
+pub mod audit_destinations;
 pub mod backup;
+pub mod backup_user;
 pub mod create_admin;
+pub mod create_user;
+pub mod delete_user;
+pub mod disable_user;
 pub mod doctor;
+pub mod enable_user;
+pub mod expand;
+pub mod list_users;
 pub mod migrate;
+pub mod repair;
+pub mod reset_password;
 pub mod restore;
+pub mod restore_user;
 pub mod serve;
+pub mod shorten;
 pub mod stats;
 pub mod validate;

 #[derive(Parser)]
 #[command(name = "bzod")]
+#[command(version)]
 #[command(about = "BZOD - Personal Redirector & Landing Page Platform")]
 pub struct Cli {
    #[command(subcommand)]
@@ -60,6 +74,11 @@ pub enum Commands {
        #[arg(long)]
        data_dir: Option<String>,
    },
+    /// Read-only audit of stored redirect destinations (schemes, control chars, malformed)
+    AuditDestinations {
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
    /// Create a new administrator user in the database
    CreateAdmin {
        #[arg(long)]
@@ -72,4 +91,119 @@ pub enum Commands {
        #[arg(long)]
        data_dir: Option<String>,
    },
+    /// Shorten a URL (Feature 3)
+    Shorten {
+        /// The destination URL to shorten
+        target_url: String,
+        /// Custom slug (starting with ! followed by a-z, 0-9, -, _)
+        #[arg(long)]
+        slug: Option<String>,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// Expand a shortened code or custom slug to its destination URL (Feature 4)
+    Expand {
+        /// The short code or custom slug to expand
+        code: String,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// Create a new standard user in the database
+    CreateUser {
+        #[arg(long)]
+        username: Option<String>,
+        #[arg(long)]
+        password: Option<String>,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// Delete a standard user and all their databases/slugs
+    DeleteUser {
+        /// User ID to delete
+        user_id: i64,
+        /// Force deletion of system account/legacy_admin
+        #[arg(long)]
+        force: bool,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// Disable a standard user
+    DisableUser {
+        /// User ID to disable
+        user_id: i64,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// Enable a standard user
+    EnableUser {
+        /// User ID to enable
+        user_id: i64,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// Reset standard user's password
+    ResetPassword {
+        /// User ID to reset
+        user_id: i64,
+        #[arg(long)]
+        password: Option<String>,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// List all standard/system users
+    ListUsers {
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// Backup a standard user's databases to a .tar.zst package
+    BackupUser {
+        /// Username to backup
+        username: String,
+        /// Output .tar.zst filepath
+        #[arg(long)]
+        out: Option<String>,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// Restore a standard user's databases from a .tar.zst package
+    RestoreUser {
+        /// Input .tar.zst package path
+        #[arg(long, required = true)]
+        file: String,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
+    /// FUTURE: Migrate legacy admin content to a specific admin tenant database
+    AdminMigrate {
+        /// Target Admin ID
+        target_admin_id: i64,
+        #[arg(long)]
+        data_dir: Option<String>,
+        /// Preview what would be moved without making changes
+        #[arg(long)]
+        dry_run: bool,
+        /// Force the migration to execute
+        #[arg(long)]
+        force: bool,
+    },
+    /// Repair registry and database inconsistencies
+    Repair {
+        #[command(subcommand)]
+        command: RepairCommands,
+    },
+}
+
+#[derive(clap::Subcommand)]
+pub enum RepairCommands {
+    /// Repair Global Slug Registry inconsistencies
+    Registry {
+        #[arg(long)]
+        dry_run: bool,
+        #[arg(long)]
+        force: bool,
+        #[arg(long)]
+        slug: Option<String>,
+        #[arg(long)]
+        data_dir: Option<String>,
+    },
 }
@@ -0,0 +1,178 @@
+use crate::cli::RepairCommands;
+use crate::config::Config;
+use crate::services::registry_validator::{RegistryIssueType, RegistryValidator};
+use rusqlite::Connection;
+use std::path::PathBuf;
+use tracing::info;
+
+pub async fn run(
+    command: RepairCommands,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    match command {
+        RepairCommands::Registry {
+            dry_run,
+            force,
+            slug,
+            data_dir,
+        } => {
+            if let Some(d) = data_dir {
+                config.data_dir = PathBuf::from(d);
+            }
+
+            if !dry_run && !force {
+                println!("Error: You must specify either --dry-run or --force");
+                return Ok(());
+            }
+            if dry_run && force {
+                println!("Error: Cannot specify both --dry-run and --force");
+                return Ok(());
+            }
+
+            let start_time = std::time::Instant::now();
+            let admin_dir = config.data_dir.join("admin");
+            let system_db_path = admin_dir.join("system.db");
+            let users_db_path = admin_dir.join("users.db");
+
+            if !system_db_path.exists() || !users_db_path.exists() {
+                println!("Error: system.db or users.db not found.");
+                return Ok(());
+            }
+
+            let mut sys_conn = Connection::open(&system_db_path)?;
+            let usr_conn = Connection::open(&users_db_path)?;
+
+            let slug_filter = slug.as_deref();
+
+            if dry_run {
+                println!("BZOD Registry Repair\n");
+                println!("Scanning Global Slug Registry...");
+
+                let issues =
+                    RegistryValidator::scan(&sys_conn, &usr_conn, &config.data_dir, slug_filter)?;
+                let orphaned = issues
+                    .into_iter()
+                    .filter(|i| {
+                        matches!(
+                            i.issue_type,
+                            RegistryIssueType::MissingTarget
+                                | RegistryIssueType::MissingDatabase
+                                | RegistryIssueType::MissingOwner
+                        )
+                    })
+                    .collect::<Vec<_>>();
+
+                let orphaned_pages = orphaned.iter().filter(|i| i.target_type == "page").count();
+                let orphaned_urls = orphaned.iter().filter(|i| i.target_type == "url").count();
+
+                println!("\nDetected:");
+                println!("\nPages:\n    {} orphaned", orphaned_pages);
+                println!("\nURLs:\n    {} orphaned", orphaned_urls);
+
+                if !orphaned.is_empty() {
+                    println!("\nThe following entries would be removed:");
+                    for issue in &orphaned {
+                        println!("\n{}\n    {}", issue.target_type.to_uppercase(), issue.slug);
+                    }
+                }
+
+                println!("\nNo changes have been made.");
+                println!(
+                    "\nRun again with:\n\n    bzod repair registry --force{}",
+                    if let Some(s) = slug_filter {
+                        format!(" --slug {}", s)
+                    } else {
+                        "".to_string()
+                    }
+                );
+
+                info!(
+                    "Registry Repair Started. Scanned. Orphaned Pages: {}, Orphaned URLs: {}. Duration: {:?}",
+                    orphaned_pages, orphaned_urls, start_time.elapsed()
+                );
+            } else if force {
+                let tx = sys_conn.transaction()?;
+
+                let issues =
+                    RegistryValidator::scan(&tx, &usr_conn, &config.data_dir, slug_filter)?;
+                let orphaned = issues
+                    .into_iter()
+                    .filter(|i| {
+                        matches!(
+                            i.issue_type,
+                            RegistryIssueType::MissingTarget
+                                | RegistryIssueType::MissingDatabase
+                                | RegistryIssueType::MissingOwner
+                        )
+                    })
+                    .collect::<Vec<_>>();
+
+                let orphaned_pages = orphaned.iter().filter(|i| i.target_type == "page").count();
+                let orphaned_urls = orphaned.iter().filter(|i| i.target_type == "url").count();
+
+                if orphaned.is_empty() {
+                    println!("No repairs required.");
+                    return Ok(());
+                }
+
+                if let Some(s) = slug_filter {
+                    println!("Checking slug:\n\n{}\n", s);
+                    if let Some(issue) = orphaned.first() {
+                        println!("Owner:\n\n{}\n", issue.owner_user_id);
+                        println!("Status:\n\nOrphaned\n");
+                    }
+                }
+
+                let mut removed_count = 0;
+                for issue in &orphaned {
+                    let rows = tx.execute(
+                        "DELETE FROM global_slugs WHERE slug = ?1",
+                        rusqlite::params![issue.slug],
+                    )?;
+                    removed_count += rows;
+                }
+
+                tx.commit()?;
+
+                if slug_filter.is_some() {
+                    println!("Removed:\n\nSUCCESS");
+                } else {
+                    println!("Repair Complete\n");
+                    println!("Removed:\n");
+                    println!("Pages:\n    {}\n", orphaned_pages);
+                    println!("URLs:\n    {}\n", orphaned_urls);
+
+                    let remaining: i64 =
+                        sys_conn
+                            .query_row("SELECT COUNT(*) FROM global_slugs;", [], |r| r.get(0))?;
+                    println!("Remaining Registry Entries:\n    {}\n", remaining);
+
+                    let post_issues =
+                        RegistryValidator::scan(&sys_conn, &usr_conn, &config.data_dir, None)?;
+                    let post_orphaned = post_issues
+                        .iter()
+                        .filter(|i| {
+                            matches!(
+                                i.issue_type,
+                                RegistryIssueType::MissingTarget
+                                    | RegistryIssueType::MissingDatabase
+                                    | RegistryIssueType::MissingOwner
+                            )
+                        })
+                        .count();
+
+                    println!(
+                        "Integrity:\n    {}",
+                        if post_orphaned == 0 { "PASS" } else { "FAIL" }
+                    );
+                }
+
+                info!(
+                    "Registry Repair Started. Scanned. Orphaned Pages: {}, Orphaned URLs: {}. Removed: {}. Duration: {:?}",
+                    orphaned_pages, orphaned_urls, removed_count, start_time.elapsed()
+                );
+            }
+        }
+    }
+    Ok(())
+}
@@ -0,0 +1,75 @@
+use crate::auth::hash_password;
+use crate::config::Config;
+use crate::db::Db;
+use std::io::{self, Write};
+use std::path::PathBuf;
+use tracing::{error, info};
+
+pub async fn run(
+    user_id: i64,
+    password: Option<String>,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    // Check user exists
+    let user = {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::get_user_by_id(&conn, user_id)?
+    };
+
+    let user = match user {
+        Some(u) => u,
+        None => {
+            error!("User ID {} not found", user_id);
+            return Ok(());
+        }
+    };
+
+    let final_password = match password {
+        Some(p) => p,
+        None => read_input("Enter new password: "),
+    };
+    if final_password.trim().is_empty() {
+        error!("Password cannot be empty");
+        return Ok(());
+    }
+
+    let hash = hash_password(&final_password).map_err(|e| e.to_string())?;
+
+    {
+        let conn = db.users.lock().unwrap();
+        crate::db::users::reset_user_password(&conn, user_id, &hash)?;
+    }
+
+    // Write audit event
+    {
+        let system_conn = db.system.lock().unwrap();
+        let _ = crate::db::audit_events::write_audit_event(
+            &system_conn,
+            "cli",
+            "USER_PASSWORD_RESET",
+            "user",
+            &user_id.to_string(),
+            Some(&format!("Username: {}", user.username)),
+        );
+    }
+
+    info!(
+        "Password for user {} (ID: {}) has been reset successfully",
+        user.username, user_id
+    );
+    Ok(())
+}
+
+fn read_input(prompt: &str) -> String {
+    print!("{}", prompt);
+    let _ = io::stdout().flush();
+    let mut input = String::new();
+    let _ = io::stdin().read_line(&mut input);
+    input.trim().to_string()
+}
@@ -1,10 +1,388 @@
 use crate::config::Config;
+use crate::services::registry_validator::RegistryIssueType;
 use flate2::read::GzDecoder;
 use std::fs::File;
 use std::io::{self, Write};
-use std::path::PathBuf;
+use std::path::{Path, PathBuf};
 use tar::Archive;
-use tracing::{error, info};
+use tracing::{error, info, warn};
+
+/// Read backup_manifest.json and return true if this is a legacy_flat_backup.
+fn is_legacy_flat_backup(temp_dir: &Path) -> bool {
+    let manifest_path = temp_dir.join("backup_manifest.json");
+    if !manifest_path.exists() {
+        return false;
+    }
+    match std::fs::read_to_string(&manifest_path) {
+        Ok(contents) => match serde_json::from_str::<serde_json::Value>(&contents) {
+            Ok(val) => val.get("type").and_then(|t| t.as_str()) == Some("legacy_flat_backup"),
+            Err(_) => false,
+        },
+        Err(_) => false,
+    }
+}
+
+/// Detect if the unpacked archive is in flat layout (files at root, not in admin/ subdirectory).
+fn is_flat_layout(temp_dir: &Path) -> bool {
+    temp_dir.join("admin.db").exists() && !temp_dir.join("admin").join("admin.db").exists()
+}
+
+/// Bootstrap users.db for a legacy backup where users.db is empty/unmigrated.
+///
+/// This function:
+/// 1. Runs USERS_MIGRATIONS on users.db to create the required schema.
+/// 2. Reads the actual administrator identity from admin.db (preserving
+///    the original username and argon2id password hash — no manufacturing).
+/// 3. Creates a legacy_admin system placeholder (id=1) for tenant ownership.
+/// 4. Creates an admin account with the original credentials.
+/// 5. Scans global_slugs for owner_user_ids and creates disabled placeholder
+///    accounts for any missing tenants.
+fn bootstrap_legacy_users_db(temp_dir: &Path) -> Result<(), Box<dyn std::error::Error>> {
+    use crate::db::migrations::{run_migrations, USERS_MIGRATIONS};
+
+    let users_db_path = temp_dir.join("admin").join("users.db");
+    let admin_db_path = temp_dir.join("admin").join("admin.db");
+    let system_db_path = temp_dir.join("admin").join("system.db");
+
+    // Check if users.db already has the users table (i.e., not a legacy backup)
+    {
+        let conn = rusqlite::Connection::open(&users_db_path)?;
+        let has_users_table: bool = conn
+            .query_row(
+                "SELECT EXISTS(SELECT 1 FROM sqlite_master WHERE type='table' AND name='users');",
+                [],
+                |r| r.get(0),
+            )
+            .unwrap_or(false);
+        if has_users_table {
+            info!("users.db already has users table; skipping legacy bootstrap");
+            return Ok(());
+        }
+    }
+
+    info!("Legacy users.db detected (empty/unmigrated). Bootstrapping current schema...");
+
+    // Step 1: Run migrations to create the users.db schema
+    let mut users_conn = rusqlite::Connection::open(&users_db_path)?;
+    crate::db::sqlite::enable_wal(&users_conn, "users")?;
+    crate::db::sqlite::enable_foreign_keys(&users_conn, "users")?;
+    run_migrations(&mut users_conn, "users", USERS_MIGRATIONS, None)?;
+
+    // Step 2: Read the actual administrator identity from admin.db
+    let (admin_username, admin_password_hash) = {
+        let admin_conn = rusqlite::Connection::open(&admin_db_path)?;
+
+        // The legacy admin.db users table has schema:
+        //   id TEXT PRIMARY KEY (UUID), username TEXT, password_hash TEXT, created_at TEXT
+        // Read the actual admin — typically the first (and often only) user.
+        let result: Result<(String, String), _> = admin_conn.query_row(
+            "SELECT username, password_hash FROM users ORDER BY created_at ASC LIMIT 1;",
+            [],
+            |row| Ok((row.get(0)?, row.get(1)?)),
+        );
+
+        match result {
+            Ok((username, hash)) => {
+                info!(
+                    "Preserved administrator identity from legacy admin.db: username='{}'",
+                    username
+                );
+                (username, hash)
+            }
+            Err(e) => {
+                return Err(format!(
+                    "Failed to read administrator credentials from legacy admin.db: {}",
+                    e
+                )
+                .into());
+            }
+        }
+    };
+
+    // Step 3: Create legacy_admin system placeholder (id=1) for tenant content ownership
+    // This account owns the content.db/analytics.db from the flat backup (users/1/).
+    // It uses the original admin's password hash so no synthetic credentials are introduced.
+    let now = chrono::Utc::now().to_rfc3339();
+    users_conn.execute(
+        "INSERT INTO users (id, username, password_hash, status, created_at, account_type)
+         VALUES (?1, ?2, ?3, ?4, ?5, ?6);",
+        rusqlite::params![
+            1i64,
+            "legacy_admin",
+            &admin_password_hash,
+            "disabled",
+            &now,
+            "system"
+        ],
+    )?;
+    users_conn.execute("INSERT INTO quotas (user_id) VALUES (?1);", [1i64])?;
+    info!("Created legacy_admin system account (id=1) for tenant content ownership");
+
+    // Step 4: Create the actual admin account with original credentials
+    users_conn.execute(
+        "INSERT INTO users (username, password_hash, status, created_at, account_type)
+         VALUES (?1, ?2, ?3, ?4, ?5);",
+        rusqlite::params![
+            &admin_username,
+            &admin_password_hash,
+            "active",
+            &now,
+            "admin"
+        ],
+    )?;
+    let admin_id = users_conn.last_insert_rowid();
+    users_conn.execute("INSERT INTO quotas (user_id) VALUES (?1);", [admin_id])?;
+    info!(
+        "Created admin account '{}' (id={}) with original credentials",
+        admin_username, admin_id
+    );
+
+    // Step 5: Scan global_slugs for owner_user_ids and create placeholders for missing tenants
+    if system_db_path.exists() {
+        let system_conn = rusqlite::Connection::open(&system_db_path)?;
+        let has_global_slugs: bool = system_conn
+            .query_row(
+                "SELECT EXISTS(SELECT 1 FROM sqlite_master WHERE type='table' AND name='global_slugs');",
+                [],
+                |r| r.get(0),
+            )
+            .unwrap_or(false);
+
+        if has_global_slugs {
+            let mut stmt =
+                system_conn.prepare("SELECT DISTINCT owner_user_id FROM global_slugs;")?;
+            let mut rows = stmt.query([])?;
+            while let Some(row) = rows.next()? {
+                let owner_id: i64 = row.get(0)?;
+                // Skip user 1 (legacy_admin) and the admin we just created
+                if owner_id == 1 || owner_id == admin_id {
+                    continue;
+                }
+
+                // Check if this user already exists in users.db
+                let exists: bool = users_conn
+                    .query_row(
+                        "SELECT EXISTS(SELECT 1 FROM users WHERE id = ?1);",
+                        [owner_id],
+                        |r| r.get(0),
+                    )
+                    .unwrap_or(false);
+
+                if !exists {
+                    // Create a disabled placeholder so RegistryValidator can resolve ownership.
+                    // The tenant's actual databases were not included in the flat backup.
+                    let placeholder_name = format!("restored_user_{}", owner_id);
+                    users_conn.execute(
+                        "INSERT INTO users (id, username, password_hash, status, created_at, account_type, metadata)
+                         VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
+                        rusqlite::params![
+                            owner_id,
+                            &placeholder_name,
+                            &admin_password_hash,
+                            "disabled",
+                            &now,
+                            "standard",
+                            "Placeholder created during legacy_flat_backup restore. Original tenant databases were not included in the flat backup."
+                        ],
+                    )?;
+                    users_conn.execute("INSERT INTO quotas (user_id) VALUES (?1);", [owner_id])?;
+                    warn!(
+                        "Created placeholder account for user_id={} (referenced in global_slugs but tenant databases not in backup)",
+                        owner_id
+                    );
+                }
+            }
+        }
+    }
+
+    Ok(())
+}
+
+/// Classify registry issues into hard errors vs warnings for legacy restore.
+///
+/// Hard errors: DuplicateSlug, InvalidTargetType, InvalidStatus
+/// Warnings:    MissingDatabase, MissingTarget, MissingOwner, StaleReservation,
+///              TenantAdminHasIsolatedContent
+fn classify_registry_issues(
+    issues: &[crate::services::registry_validator::RegistryIssue],
+    is_legacy: bool,
+) -> (
+    Vec<&crate::services::registry_validator::RegistryIssue>,
+    Vec<&crate::services::registry_validator::RegistryIssue>,
+) {
+    let mut errors = Vec::new();
+    let mut warnings = Vec::new();
+
+    for issue in issues {
+        match issue.issue_type {
+            RegistryIssueType::DuplicateSlug
+            | RegistryIssueType::InvalidTargetType
+            | RegistryIssueType::InvalidStatus => {
+                errors.push(issue);
+            }
+            RegistryIssueType::MissingOwner if !is_legacy => {
+                errors.push(issue);
+            }
+            _ => {
+                // For legacy restores: MissingDatabase, MissingTarget, MissingOwner,
+                // StaleReservation, TenantAdminHasIsolatedContent are warnings.
+                // These represent pre-existing inconsistencies in the backup data,
+                // not restore corruption.
+                warnings.push(issue);
+            }
+        }
+    }
+
+    (errors, warnings)
+}
+
+pub fn perform_restore(
+    file_path: &Path,
+    data_dir: &Path,
+) -> Result<(), Box<dyn std::error::Error>> {
+    // 1. Open and unpack the archive to a temporary directory
+    let f = File::open(file_path)?;
+    let tar_gz = GzDecoder::new(f);
+    let mut archive = Archive::new(tar_gz);
+
+    let temp_dir =
+        std::env::temp_dir().join(format!("bzod_system_restore_{}", uuid::Uuid::new_v4()));
+    std::fs::create_dir_all(&temp_dir)?;
+
+    if let Err(e) = archive.unpack(&temp_dir) {
+        let _ = std::fs::remove_dir_all(&temp_dir);
+        return Err(e.into());
+    }
+
+    // 2. Detect backup format
+    let is_legacy = is_legacy_flat_backup(&temp_dir);
+    let needs_normalization = is_flat_layout(&temp_dir);
+
+    if is_legacy {
+        info!("Detected legacy_flat_backup format — using legacy-aware restore path");
+    }
+
+    // 3. Normalize flat layout into multi-tenant structure BEFORE any validation
+    if needs_normalization {
+        info!("Normalizing flat database layout into multi-tenant structure...");
+        if let Err(e) = crate::services::backup_layout::normalize_restored_layout(&temp_dir) {
+            let _ = std::fs::remove_dir_all(&temp_dir);
+            return Err(format!("Failed to normalize legacy layout: {}", e).into());
+        }
+    }
+
+    // 4. For legacy backups: bootstrap the empty users.db with the current schema
+    //    and populate it from admin.db credentials
+    if is_legacy {
+        if let Err(e) = bootstrap_legacy_users_db(&temp_dir) {
+            let _ = std::fs::remove_dir_all(&temp_dir);
+            return Err(format!("Failed to bootstrap legacy users database: {}", e).into());
+        }
+    }
+
+    // 5. Run validation on the normalized temp_dir
+    let mut temp_config = Config::load();
+    temp_config.data_dir = temp_dir.clone();
+
+    // Namespace audit
+    match crate::db::users::audit_slug_namespace(&temp_config) {
+        Ok(report) => {
+            if !report.duplicates.is_empty() {
+                let _ = std::fs::remove_dir_all(&temp_dir);
+                return Err(
+                    format!("Slug conflicts detected in backup: {:?}", report.duplicates).into(),
+                );
+            }
+        }
+        Err(e) => {
+            let _ = std::fs::remove_dir_all(&temp_dir);
+            return Err(format!("Failed to audit slug namespace in backup: {}", e).into());
+        }
+    }
+
+    // Registry integrity check
+    let system_db_path = temp_dir.join("admin").join("system.db");
+    let users_db_path = temp_dir.join("admin").join("users.db");
+
+    if system_db_path.exists() && users_db_path.exists() {
+        let system_conn = rusqlite::Connection::open(&system_db_path)?;
+        let users_conn = rusqlite::Connection::open(&users_db_path)?;
+        match crate::services::registry_validator::RegistryValidator::scan(
+            &system_conn,
+            &users_conn,
+            &temp_dir,
+            None,
+        ) {
+            Ok(issues) => {
+                if !issues.is_empty() {
+                    let (hard_errors, warnings) = classify_registry_issues(&issues, is_legacy);
+
+                    // Log all warnings
+                    for w in &warnings {
+                        warn!(
+                            "Legacy restore warning: {:?} — {}",
+                            w.issue_type, w.description
+                        );
+                    }
+
+                    // Abort only on hard errors
+                    if !hard_errors.is_empty() {
+                        let descriptions: Vec<String> = hard_errors
+                            .iter()
+                            .map(|e| format!("{:?}: {}", e.issue_type, e.description))
+                            .collect();
+                        let _ = std::fs::remove_dir_all(&temp_dir);
+                        return Err(format!(
+                            "Registry integrity errors in backup ({} critical): {}",
+                            hard_errors.len(),
+                            descriptions.join("; ")
+                        )
+                        .into());
+                    }
+
+                    if !warnings.is_empty() {
+                        info!(
+                            "Registry validation completed with {} warnings (pre-existing backup inconsistencies)",
+                            warnings.len()
+                        );
+                    }
+                }
+            }
+            Err(e) => {
+                let _ = std::fs::remove_dir_all(&temp_dir);
+                return Err(format!("Failed to verify registry integrity in backup: {}", e).into());
+            }
+        }
+    }
+
+    // 6. If validation succeeds, atomically replace data_dir contents
+    if data_dir.exists() {
+        let _ = std::fs::remove_dir_all(data_dir);
+    }
+    std::fs::create_dir_all(data_dir)?;
+
+    fn copy_dir_all(src: &Path, dst: &Path) -> std::io::Result<()> {
+        std::fs::create_dir_all(dst)?;
+        for entry in std::fs::read_dir(src)? {
+            let entry = entry?;
+            let ty = entry.file_type()?;
+            if ty.is_dir() {
+                copy_dir_all(&entry.path(), &dst.join(entry.file_name()))?;
+            } else {
+                std::fs::copy(entry.path(), dst.join(entry.file_name()))?;
+            }
+        }
+        Ok(())
+    }
+
+    if let Err(e) = copy_dir_all(&temp_dir, data_dir) {
+        let _ = std::fs::remove_dir_all(&temp_dir);
+        return Err(format!("Failed to copy restored files: {}", e).into());
+    }
+
+    let _ = std::fs::remove_dir_all(&temp_dir);
+    Ok(())
+}

 pub async fn run(
    file: String,
@@ -40,10 +418,7 @@ pub async fn run(
    }

    info!("Restoring backup from: {:?}", file_path);
-    let f = File::open(&file_path)?;
-    let tar_gz = GzDecoder::new(f);
-    let mut archive = Archive::new(tar_gz);
-    archive.unpack(&config.data_dir)?;
+    perform_restore(&file_path, &config.data_dir)?;
    info!("Database files successfully restored.");

    Ok(())
@@ -0,0 +1,182 @@
+use crate::config::Config;
+use crate::db::Db;
+use std::fs::File;
+use std::path::PathBuf;
+use tar::Archive;
+use tracing::{error, info};
+use zstd::Decoder;
+
+#[derive(serde::Serialize, serde::Deserialize)]
+struct UserBackupMetadata {
+    id: i64,
+    username: String,
+    password_hash: String,
+    status: String,
+    created_at: String,
+    account_type: String,
+    metadata: Option<String>,
+    quotas: UserBackupQuotas,
+}
+
+#[derive(serde::Serialize, serde::Deserialize)]
+struct UserBackupQuotas {
+    max_urls: i64,
+    max_landings: i64,
+    max_api_tokens: i64,
+    max_storage_mb: i64,
+}
+
+pub async fn run(
+    file: String,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let file_path = PathBuf::from(file);
+
+    if !file_path.exists() {
+        error!("Backup file not found: {:?}", file_path);
+        return Ok(());
+    }
+
+    let db = Db::init(&config)?;
+
+    // 1. Read metadata.json from the tar.zst archive
+    let f = File::open(&file_path)?;
+    let zst_dec = Decoder::new(f)?;
+    let mut archive = Archive::new(zst_dec);
+
+    let mut metadata_opt: Option<UserBackupMetadata> = None;
+    for entry_res in archive.entries()? {
+        let mut entry = entry_res?;
+        let path = entry.path()?;
+        let file_name = path.file_name().and_then(|n| n.to_str()).unwrap_or("");
+        if file_name == "metadata.json" {
+            let meta: UserBackupMetadata = serde_json::from_reader(&mut entry)?;
+            metadata_opt = Some(meta);
+            break;
+        }
+    }
+
+    let metadata = match metadata_opt {
+        Some(m) => m,
+        None => {
+            error!("Archive is missing metadata.json");
+            return Ok(());
+        }
+    };
+
+    info!("Restoring user {} from backup...", metadata.username);
+
+    // 2. Resolve target user ID and upsert user record in users.db
+    let target_user_id = {
+        let users_conn = db.users.lock().unwrap();
+        let existing_user =
+            crate::db::users::get_user_by_username(&users_conn, &metadata.username)?;
+
+        match existing_user {
+            Some(u) => {
+                users_conn.execute(
+                    "UPDATE users SET password_hash = ?1, status = ?2, account_type = ?3, metadata = ?4 WHERE id = ?5;",
+                    rusqlite::params![metadata.password_hash, metadata.status, metadata.account_type, metadata.metadata, u.id],
+                )?;
+                users_conn.execute(
+                    "INSERT OR REPLACE INTO quotas (user_id, max_urls, max_landings, max_api_tokens, max_storage_mb) 
+                     VALUES (?1, ?2, ?3, ?4, ?5);",
+                    rusqlite::params![u.id, metadata.quotas.max_urls, metadata.quotas.max_landings, metadata.quotas.max_api_tokens, metadata.quotas.max_storage_mb],
+                )?;
+                u.id
+            }
+            None => {
+                let id_taken: bool = users_conn
+                    .query_row(
+                        "SELECT EXISTS(SELECT 1 FROM users WHERE id = ?1);",
+                        [metadata.id],
+                        |row| row.get(0),
+                    )
+                    .unwrap_or(false);
+
+                let new_id = if !id_taken {
+                    users_conn.execute(
+                        "INSERT INTO users (id, username, password_hash, status, created_at, account_type, metadata) 
+                         VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
+                        rusqlite::params![metadata.id, metadata.username, metadata.password_hash, metadata.status, metadata.created_at, metadata.account_type, metadata.metadata],
+                    )?;
+                    metadata.id
+                } else {
+                    users_conn.execute(
+                        "INSERT INTO users (username, password_hash, status, created_at, account_type, metadata) 
+                         VALUES (?1, ?2, ?3, ?4, ?5, ?6);",
+                        rusqlite::params![metadata.username, metadata.password_hash, metadata.status, metadata.created_at, metadata.account_type, metadata.metadata],
+                    )?;
+                    users_conn.last_insert_rowid()
+                };
+
+                users_conn.execute(
+                    "INSERT OR REPLACE INTO quotas (user_id, max_urls, max_landings, max_api_tokens, max_storage_mb) 
+                     VALUES (?1, ?2, ?3, ?4, ?5);",
+                    rusqlite::params![new_id, metadata.quotas.max_urls, metadata.quotas.max_landings, metadata.quotas.max_api_tokens, metadata.quotas.max_storage_mb],
+                )?;
+                new_id
+            }
+        }
+    };
+
+    // 3. Extract database files to /data/users/<target_user_id>/
+    let dest_dir = config
+        .data_dir
+        .join("users")
+        .join(target_user_id.to_string());
+    std::fs::create_dir_all(&dest_dir)?;
+
+    let f2 = File::open(&file_path)?;
+    let zst_dec2 = Decoder::new(f2)?;
+    let mut archive2 = Archive::new(zst_dec2);
+
+    for entry_res in archive2.entries()? {
+        let mut entry = entry_res?;
+        let path = entry.path()?;
+        let file_name = path.file_name().and_then(|n| n.to_str()).unwrap_or("");
+        match file_name {
+            "content.db" => {
+                let mut out_file = File::create(dest_dir.join("content.db"))?;
+                std::io::copy(&mut entry, &mut out_file)?;
+            }
+            "analytics.db" => {
+                let mut out_file = File::create(dest_dir.join("analytics.db"))?;
+                std::io::copy(&mut entry, &mut out_file)?;
+            }
+            "profile.db" => {
+                let mut out_file = File::create(dest_dir.join("profile.db"))?;
+                std::io::copy(&mut entry, &mut out_file)?;
+            }
+            _ => {}
+        }
+    }
+
+    // 4. Register slugs in global_slugs using the shared helper
+    {
+        let system_conn = db.system.lock().unwrap();
+        crate::db::users::register_restored_user_slugs(
+            &system_conn,
+            target_user_id,
+            &dest_dir.join("content.db"),
+        )?;
+    }
+
+    // 5. Reconcile quotas for restored user
+    let restored_content_conn = rusqlite::Connection::open(dest_dir.join("content.db"))?;
+    crate::db::users::reconcile_user_quotas(
+        &db.users.lock().unwrap(),
+        target_user_id,
+        &restored_content_conn,
+    )?;
+
+    info!(
+        "User '{}' (ID: {}) successfully restored from backup.",
+        metadata.username, target_user_id
+    );
+    Ok(())
+}
@@ -7,6 +7,30 @@ use std::path::PathBuf;
 use std::time::Instant;
 use tracing::info;

+async fn shutdown_signal() {
+    let ctrl_c = async {
+        tokio::signal::ctrl_c()
+            .await
+            .expect("failed to install Ctrl+C handler");
+    };
+
+    #[cfg(unix)]
+    let terminate = async {
+        tokio::signal::unix::signal(tokio::signal::unix::SignalKind::terminate())
+            .expect("failed to install signal handler")
+            .recv()
+            .await;
+    };
+
+    #[cfg(not(unix))]
+    let terminate = std::future::pending::<()>();
+
+    tokio::select! {
+        _ = ctrl_c => {},
+        _ = terminate => {},
+    }
+}
+
 pub async fn run(
    host: Option<String>,
    port: Option<u16>,
@@ -29,45 +53,91 @@ pub async fn run(
    // Init DBs
    let db = Db::init(&config)?;

+    let (shutdown_tx, shutdown_rx) = tokio::sync::watch::channel(false);
+    let mut join_handles = Vec::new();
+
    // Init Queue
-    let queue = AnalyticsQueue::new(db.clone(), 1000);
+    let (queue, analytics_handle) = AnalyticsQueue::new(db.clone(), 1000, shutdown_rx.clone());
+    join_handles.push(("analytics_worker", analytics_handle));

    // Spawn background tasks
    let link_checker_db = db.clone();
    let link_checker_interval = config.link_check_interval_mins;
-    tokio::spawn(async move {
-        crate::jobs::run_link_checker(link_checker_db, link_checker_interval).await;
-    });
+    let rx = shutdown_rx.clone();
+    join_handles.push((
+        "link_checker",
+        tokio::spawn(async move {
+            crate::jobs::run_link_checker(link_checker_db, link_checker_interval, rx).await;
+        }),
+    ));

    let aggregator_db = db.clone();
    let aggregator_interval = config.aggregation_interval_mins;
-    tokio::spawn(async move {
-        crate::jobs::run_aggregator(aggregator_db, aggregator_interval).await;
-    });
+    let rx = shutdown_rx.clone();
+    join_handles.push((
+        "aggregator",
+        tokio::spawn(async move {
+            crate::jobs::run_aggregator(aggregator_db, aggregator_interval, rx).await;
+        }),
+    ));

    let retention_db = db.clone();
    let retention_days = config.data_retention_days;
-    tokio::spawn(async move {
-        crate::jobs::run_retention_cleaner(retention_db, retention_days).await;
-    });
+    let rx = shutdown_rx.clone();
+    join_handles.push((
+        "retention_cleaner",
+        tokio::spawn(async move {
+            crate::jobs::run_retention_cleaner(retention_db, retention_days, rx).await;
+        }),
+    ));

    // Spawn optional backup scheduler
    let backup_db = db.clone();
    let backup_config = config.clone();
-    tokio::spawn(async move {
-        crate::jobs::backup::run_backup_scheduler(backup_db, backup_config).await;
-    });
+    let rx = shutdown_rx.clone();
+    join_handles.push((
+        "backup_scheduler",
+        tokio::spawn(async move {
+            crate::jobs::backup::run_backup_scheduler(backup_db, backup_config, rx).await;
+        }),
+    ));

    let expiry_db = db.clone();
-    tokio::spawn(async move {
-        crate::jobs::run_expiry_checker(expiry_db).await;
-    });
+    let rx = shutdown_rx.clone();
+    join_handles.push((
+        "expiry_checker",
+        tokio::spawn(async move {
+            crate::jobs::run_expiry_checker(expiry_db, rx).await;
+        }),
+    ));
+
+    let reconcile_db = db.clone();
+    let reconcile_interval_hours = {
+        let conn = db.system.lock().unwrap();
+        conn.query_row(
+            "SELECT value FROM settings WHERE key = 'quota_reconcile_interval_hours';",
+            [],
+            |row| row.get::<_, String>(0),
+        )
+        .ok()
+        .and_then(|val| val.parse::<u64>().ok())
+        .unwrap_or(24)
+    };
+    let rx = shutdown_rx.clone();
+    join_handles.push((
+        "quota_reconciliation",
+        tokio::spawn(async move {
+            crate::jobs::run_quota_reconciliation(reconcile_db, reconcile_interval_hours, rx).await;
+        }),
+    ));

    let state = AppState {
        admin_db: db.admin.clone(),
        content_db: db.content.clone(),
        analytics_db: db.analytics.clone(),
        system_db: db.system.clone(),
+        users_db: db.users.clone(),
+        user_dbs: std::sync::Arc::new(std::sync::Mutex::new(std::collections::HashMap::new())),
        db: db.clone(),
        config: config.clone(),
        analytics_queue: queue,
@@ -80,7 +150,31 @@ pub async fn run(
    let listener = tokio::net::TcpListener::bind(&addr).await?;

    info!("Listening for requests on http://{}", addr);
-    axum::serve(listener, router).await?;
+
+    axum::serve(listener, router)
+        .with_graceful_shutdown(async move {
+            shutdown_signal().await;
+            info!("Shutdown signal received");
+            info!("Stopping HTTP server...");
+            let _ = shutdown_tx.send(true);
+        })
+        .await?;
+
+    info!("Stopping background workers...");
+
+    let timeout_duration = std::time::Duration::from_secs(10);
+    let deadline = tokio::time::Instant::now() + timeout_duration;
+
+    for (name, handle) in join_handles {
+        match tokio::time::timeout_at(deadline, handle).await {
+            Ok(Ok(_)) => {}
+            Ok(Err(e)) => tracing::error!("Background task '{}' panicked: {:?}", name, e),
+            Err(_) => tracing::warn!("Background task did not terminate: {}", name),
+        }
+    }
+
+    info!("Background workers stopped");
+    info!("BZOD shutdown complete");

    Ok(())
 }
@@ -0,0 +1,95 @@
+use crate::config::Config;
+use crate::db::Db;
+use std::path::PathBuf;
+
+pub async fn run(
+    target_url: String,
+    slug: Option<String>,
+    data_dir: Option<String>,
+    mut config: Config,
+) -> Result<(), Box<dyn std::error::Error>> {
+    // 1. Basic URL validation
+    if reqwest::Url::parse(&target_url).is_err() {
+        return Err("Invalid destination URL format".into());
+    }
+
+    if let Some(d) = data_dir {
+        config.data_dir = PathBuf::from(d);
+    }
+    let db = Db::init(&config)?;
+
+    // 2. Validate/normalize slug/code
+    let code = match slug {
+        Some(s) => {
+            let normalized = s.trim().to_lowercase();
+            if !crate::utils::validation::validate_custom_slug(&normalized) {
+                return Err(
+                    "Custom slug must start with ! followed by 1-24 characters of a-z, 0-9, -, _"
+                        .into(),
+                );
+            }
+            normalized
+        }
+        None => crate::utils::random::generate_token(3),
+    };
+
+    // 3. Register slug in system.db with status 'reserving' and check availability
+    {
+        let system_conn = db.system.lock().unwrap();
+        if !crate::db::users::is_slug_available(&system_conn, &code)? {
+            return Err("Short code/slug already exists".into());
+        }
+        crate::db::users::register_global_slug(&system_conn, &code, 1, "url", "", "reserving")?;
+    }
+
+    // 4. Persist URL
+    let conn = db.content.lock().unwrap();
+    let res = crate::db::content::create_url_extended(
+        &conn,
+        &code,
+        &target_url,
+        None,
+        None,
+        &[],
+        None,
+        None,
+        None,
+    );
+
+    match res {
+        Ok(url) => {
+            // Activate slug in system.db
+            {
+                let system_conn = db.system.lock().unwrap();
+                system_conn.execute(
+                    "UPDATE global_slugs SET target_id = ?1, status = 'active', updated_at = ?2 WHERE slug = ?3;",
+                    rusqlite::params![url.id, chrono::Utc::now().to_rfc3339(), code],
+                )?;
+            }
+            // Increment quota for user ID 1
+            {
+                let users_conn = db.users.lock().unwrap();
+                crate::db::users::increment_quota_counter(&users_conn, 1, "urls")?;
+            }
+
+            let proto = if config.cookie_secure {
+                "https"
+            } else {
+                "http"
+            };
+            let base_url = config
+                .base_url
+                .clone()
+                .unwrap_or_else(|| format!("{}://localhost:{}", proto, config.port));
+
+            // Output only the shortened URL as requested
+            println!("{}/{}", base_url, code);
+            Ok(())
+        }
+        Err(e) => {
+            let system_conn = db.system.lock().unwrap();
+            let _ = crate::db::users::release_global_slug(&system_conn, &code, 1);
+            Err(e.into())
+        }
+    }
+}
@@ -14,14 +14,26 @@ pub async fn run(
    println!("=== BZOD Database Stats ===");
    println!("Storage Directory: {:?}", config.data_dir);

-    let files = vec!["admin.db", "content.db", "analytics.db", "system.db"];
-    for f in files {
-        let p = config.data_dir.join(f);
-        if p.exists() {
-            let sz = std::fs::metadata(&p)?.len();
+    let files = vec![
+        ("admin.db", config.data_dir.join("admin/admin.db")),
+        ("system.db", config.data_dir.join("admin/system.db")),
+        ("users.db", config.data_dir.join("admin/users.db")),
+        (
+            "legacy content.db",
+            config.data_dir.join("users/1/content.db"),
+        ),
+        (
+            "legacy analytics.db",
+            config.data_dir.join("users/1/analytics.db"),
+        ),
+    ];
+
+    for (name, path) in files {
+        if path.exists() {
+            let sz = std::fs::metadata(&path)?.len();
            println!(
                "  File: {} - Size: {} bytes ({:.2} MB)",
-                f,
+                name,
                sz,
                sz as f64 / 1_048_576.0
            );
@@ -71,10 +71,20 @@ pub fn create_session(
 ) -> rusqlite::Result<Session> {
    let created_at = Utc::now().to_rfc3339();

-    conn.execute(
-        "INSERT INTO sessions (id, user_id, expires_at, created_at) VALUES (?1, ?2, ?3, ?4);",
-        params![session_id, user_id, expires_at_rfc3339, created_at],
-    )?;
+    // Bind `user_id` as integer when it appears to be numeric so that numeric
+    // user IDs inserted into `users.db` keep the integer affinity and avoid
+    // InvalidColumnType errors when read as i64 elsewhere.
+    if let Ok(id_i64) = user_id.parse::<i64>() {
+        conn.execute(
+            "INSERT INTO sessions (id, user_id, expires_at, created_at) VALUES (?1, ?2, ?3, ?4);",
+            params![session_id, id_i64, expires_at_rfc3339, created_at],
+        )?;
+    } else {
+        conn.execute(
+            "INSERT INTO sessions (id, user_id, expires_at, created_at) VALUES (?1, ?2, ?3, ?4);",
+            params![session_id, user_id, expires_at_rfc3339, created_at],
+        )?;
+    }

    Ok(Session {
        id: session_id.to_string(),
@@ -82,8 +82,8 @@ pub fn insert_visits_batch(conn: &mut Connection, records: &[VisitRecord]) -> ru
    let tx = conn.transaction()?;
    {
        let mut stmt = tx.prepare(
-            "INSERT INTO visits (id, target_type, target_id, timestamp, ip_address, user_agent, referer, accept_language, country, status_code)
-             VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7, ?8, ?9, ?10);"
+            "INSERT INTO visits (id, target_type, target_id, timestamp, ip_address, user_agent, referer, accept_language, country, status_code, owner_user_id)
+             VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7, ?8, ?9, ?10, ?11);"
        )?;

        for r in records {
@@ -97,7 +97,8 @@ pub fn insert_visits_batch(conn: &mut Connection, records: &[VisitRecord]) -> ru
                r.referer,
                r.accept_language,
                r.country,
-                r.status_code
+                r.status_code,
+                r.owner_user_id
            ])?;
        }
    }
@@ -508,6 +509,243 @@ pub fn get_metric_rankings_raw(
    Ok(res)
 }

+/// Returns the raw visit counts for a specific target ID.
+pub fn get_target_visit_count(
+    conn: &Connection,
+    target_type: &str,
+    target_id: &str,
+) -> rusqlite::Result<i64> {
+    conn.query_row(
+        "SELECT COUNT(*) FROM visits WHERE target_type = ?1 AND target_id = ?2;",
+        params![target_type, target_id],
+        |row| row.get(0),
+    )
+}
+
+/// Returns the distinct IP count (Unique Visitors) for a specific target ID.
+pub fn get_target_unique_visitors(
+    conn: &Connection,
+    target_type: &str,
+    target_id: &str,
+) -> rusqlite::Result<i64> {
+    conn.query_row(
+        "SELECT COUNT(DISTINCT ip_address) FROM visits WHERE target_type = ?1 AND target_id = ?2;",
+        params![target_type, target_id],
+        |row| row.get(0),
+    )
+}
+
+/// Fetches the monthly clicks trend for a specific target ID, falling back to a raw visits query if monthly_summaries are empty.
+pub fn get_monthly_clicks_trend(
+    conn: &Connection,
+    target_type: &str,
+    target_id: &str,
+    limit_months: i64,
+) -> rusqlite::Result<Vec<(String, i64)>> {
+    let mut stmt = conn.prepare(
+        "SELECT year_month, SUM(metric_value) FROM monthly_summaries
+         WHERE target_type = ?1 AND target_id = ?2 AND metric_type = 'clicks'
+         GROUP BY year_month ORDER BY year_month ASC LIMIT ?3;",
+    )?;
+    let rows = stmt.query_map(params![target_type, target_id, limit_months], |row| {
+        Ok((row.get::<_, String>(0)?, row.get::<_, i64>(1)?))
+    })?;
+    let mut res = Vec::new();
+    for r in rows {
+        res.push(r?);
+    }
+
+    if res.is_empty() {
+        // Fallback to raw visits
+        let mut stmt = conn.prepare(
+            "SELECT strftime('%Y-%m', timestamp) as m, COUNT(*) FROM visits
+             WHERE target_type = ?1 AND target_id = ?2
+             GROUP BY m ORDER BY m ASC LIMIT ?3;",
+        )?;
+        let rows = stmt.query_map(params![target_type, target_id, limit_months], |row| {
+            Ok((row.get::<_, String>(0)?, row.get::<_, i64>(1)?))
+        })?;
+        for r in rows {
+            res.push(r?);
+        }
+    }
+
+    Ok(res)
+}
+
+pub fn get_visits_schema_columns(
+    conn: &Connection,
+) -> rusqlite::Result<std::collections::HashSet<String>> {
+    let mut columns = std::collections::HashSet::new();
+    let mut stmt = conn.prepare("PRAGMA table_info(visits);")?;
+    let mut rows = stmt.query([])?;
+    while let Some(row) = rows.next()? {
+        let name: String = row.get("name")?;
+        columns.insert(name);
+    }
+    Ok(columns)
+}
+
+pub fn get_target_visits_paginated(
+    conn: &Connection,
+    target_type: &str,
+    target_id: &str,
+    limit: i64,
+    offset: i64,
+    date_from: Option<&str>,
+    date_to: Option<&str>,
+) -> rusqlite::Result<Vec<VisitRecord>> {
+    let mut sql = "SELECT id, target_type, target_id, timestamp, ip_address, user_agent, referer, accept_language, country, status_code, owner_user_id FROM visits WHERE target_type = ?1 AND target_id = ?2".to_string();
+    let mut params: Vec<Box<dyn rusqlite::ToSql>> = vec![
+        Box::new(target_type.to_string()),
+        Box::new(target_id.to_string()),
+    ];
+
+    if let Some(df) = date_from {
+        sql.push_str(&format!(" AND timestamp >= ?{}", params.len() + 1));
+        params.push(Box::new(format!("{}T00:00:00Z", df)));
+    }
+
+    if let Some(dt) = date_to {
+        if let Ok(parsed_date) = chrono::NaiveDate::parse_from_str(dt, "%Y-%m-%d") {
+            let next_day = parsed_date + chrono::Duration::days(1);
+            sql.push_str(&format!(" AND timestamp < ?{}", params.len() + 1));
+            params.push(Box::new(format!(
+                "{}T00:00:00Z",
+                next_day.format("%Y-%m-%d")
+            )));
+        }
+    }
+
+    sql.push_str(" ORDER BY timestamp DESC, id DESC LIMIT ?");
+    sql.push_str(&(params.len() + 1).to_string());
+    params.push(Box::new(limit));
+
+    sql.push_str(" OFFSET ?");
+    sql.push_str(&(params.len() + 1).to_string());
+    params.push(Box::new(offset));
+
+    let mut stmt = conn.prepare(&sql)?;
+    let param_refs: Vec<&dyn rusqlite::ToSql> = params.iter().map(|p| p.as_ref()).collect();
+    let rows = stmt.query_map(rusqlite::params_from_iter(param_refs), |row| {
+        Ok(VisitRecord {
+            id: row.get("id")?,
+            target_type: row.get("target_type")?,
+            target_id: row.get("target_id")?,
+            timestamp: row.get("timestamp")?,
+            ip_address: row.get("ip_address")?,
+            user_agent: row.get("user_agent")?,
+            referer: row.get("referer")?,
+            accept_language: row.get("accept_language")?,
+            country: row.get("country")?,
+            status_code: row.get("status_code")?,
+            owner_user_id: row.get("owner_user_id")?,
+        })
+    })?;
+
+    let mut visits = Vec::new();
+    for r in rows {
+        visits.push(r?);
+    }
+    Ok(visits)
+}
+
+pub fn get_target_visits_all_in_memory(
+    conn: &Connection,
+    target_type: &str,
+    target_id: &str,
+    date_from: Option<&str>,
+    date_to: Option<&str>,
+) -> rusqlite::Result<Vec<VisitRecord>> {
+    let mut sql = "SELECT id, target_type, target_id, timestamp, ip_address, user_agent, referer, accept_language, country, status_code, owner_user_id FROM visits WHERE target_type = ?1 AND target_id = ?2".to_string();
+    let mut params: Vec<Box<dyn rusqlite::ToSql>> = vec![
+        Box::new(target_type.to_string()),
+        Box::new(target_id.to_string()),
+    ];
+
+    if let Some(df) = date_from {
+        sql.push_str(&format!(" AND timestamp >= ?{}", params.len() + 1));
+        params.push(Box::new(format!("{}T00:00:00Z", df)));
+    }
+
+    if let Some(dt) = date_to {
+        if let Ok(parsed_date) = chrono::NaiveDate::parse_from_str(dt, "%Y-%m-%d") {
+            let next_day = parsed_date + chrono::Duration::days(1);
+            sql.push_str(&format!(" AND timestamp < ?{}", params.len() + 1));
+            params.push(Box::new(format!(
+                "{}T00:00:00Z",
+                next_day.format("%Y-%m-%d")
+            )));
+        }
+    }
+
+    sql.push_str(" ORDER BY timestamp DESC, id DESC");
+
+    let mut stmt = conn.prepare(&sql)?;
+    let param_refs: Vec<&dyn rusqlite::ToSql> = params.iter().map(|p| p.as_ref()).collect();
+    let rows = stmt.query_map(rusqlite::params_from_iter(param_refs), |row| {
+        Ok(VisitRecord {
+            id: row.get("id")?,
+            target_type: row.get("target_type")?,
+            target_id: row.get("target_id")?,
+            timestamp: row.get("timestamp")?,
+            ip_address: row.get("ip_address")?,
+            user_agent: row.get("user_agent")?,
+            referer: row.get("referer")?,
+            accept_language: row.get("accept_language")?,
+            country: row.get("country")?,
+            status_code: row.get("status_code")?,
+            owner_user_id: row.get("owner_user_id")?,
+        })
+    })?;
+
+    let mut visits = Vec::new();
+    for r in rows {
+        visits.push(r?);
+    }
+    Ok(visits)
+}
+
+pub fn get_target_visit_total_filtered(
+    conn: &Connection,
+    target_type: &str,
+    target_id: &str,
+    date_from: Option<&str>,
+    date_to: Option<&str>,
+) -> rusqlite::Result<i64> {
+    if date_from.is_none() && date_to.is_none() {
+        return get_target_visit_count(conn, target_type, target_id);
+    }
+
+    let mut sql =
+        "SELECT COUNT(*) FROM visits WHERE target_type = ?1 AND target_id = ?2".to_string();
+    let mut params: Vec<Box<dyn rusqlite::ToSql>> = vec![
+        Box::new(target_type.to_string()),
+        Box::new(target_id.to_string()),
+    ];
+
+    if let Some(df) = date_from {
+        sql.push_str(&format!(" AND timestamp >= ?{}", params.len() + 1));
+        params.push(Box::new(format!("{}T00:00:00Z", df)));
+    }
+
+    if let Some(dt) = date_to {
+        if let Ok(parsed_date) = chrono::NaiveDate::parse_from_str(dt, "%Y-%m-%d") {
+            let next_day = parsed_date + chrono::Duration::days(1);
+            sql.push_str(&format!(" AND timestamp < ?{}", params.len() + 1));
+            params.push(Box::new(format!(
+                "{}T00:00:00Z",
+                next_day.format("%Y-%m-%d")
+            )));
+        }
+    }
+
+    let param_refs: Vec<&dyn rusqlite::ToSql> = params.iter().map(|p| p.as_ref()).collect();
+    conn.query_row(&sql, rusqlite::params_from_iter(param_refs), |row| {
+        row.get(0)
+    })
+}
+
 #[cfg(test)]
 mod tests {
    use super::*;
@@ -47,6 +47,19 @@ pub fn get_tags_for_url(conn: &Connection, url_id: &str) -> rusqlite::Result<Vec
    Ok(tags)
 }

+/// Get the total count of URLs associated with a specific tag name.
+pub fn get_url_count_by_tag(conn: &Connection, tag: &str) -> rusqlite::Result<i64> {
+    let tag_name = tag.trim().to_lowercase();
+    conn.query_row(
+        "SELECT COUNT(*) FROM urls u
+         JOIN url_tags ut ON u.id = ut.url_id
+         JOIN tags t ON ut.tag_id = t.id
+         WHERE t.name = ?1;",
+        params![tag_name],
+        |row| row.get(0),
+    )
+}
+
 /// The full column list used in all URL SELECT queries.
 const URL_COLUMNS: &str = "id, code, destination, title, description, status, created_at, updated_at, expires_at, expired, password_hash, last_status, last_latency_ms, max_access_count, access_count";

@@ -111,10 +111,11 @@ pub fn print_migration_plan(
 // Migration definitions
 // ---------------------------------------------------------------------------

-pub const ADMIN_MIGRATIONS: &[Migration] = &[Migration {
-    version: 1,
-    name: "initial_schema",
-    sql: r#"
+pub const ADMIN_MIGRATIONS: &[Migration] = &[
+    Migration {
+        version: 1,
+        name: "initial_schema",
+        sql: r#"
    CREATE TABLE IF NOT EXISTS users (
        id TEXT PRIMARY KEY,
        username TEXT NOT NULL UNIQUE,
@@ -156,7 +157,26 @@ pub const ADMIN_MIGRATIONS: &[Migration] = &[Migration {
        value TEXT NOT NULL
    );
    "#,
-}];
+    },
+    Migration {
+        version: 2,
+        name: "remove_api_keys_fk",
+        sql: r#"
+    CREATE TABLE api_keys_new (
+        id TEXT PRIMARY KEY,
+        user_id TEXT NOT NULL,
+        key_hash TEXT NOT NULL UNIQUE,
+        name TEXT NOT NULL,
+        created_at TEXT NOT NULL,
+        last_used_at TEXT
+    );
+    INSERT INTO api_keys_new (id, user_id, key_hash, name, created_at, last_used_at)
+    SELECT id, user_id, key_hash, name, created_at, last_used_at FROM api_keys;
+    DROP TABLE api_keys;
+    ALTER TABLE api_keys_new RENAME TO api_keys;
+    "#,
+    },
+];

 pub const CONTENT_MIGRATIONS: &[Migration] = &[
    Migration {
@@ -315,6 +335,11 @@ pub const ANALYTICS_MIGRATIONS: &[Migration] = &[
    CREATE INDEX IF NOT EXISTS idx_qr_access_ts ON qr_access_log(timestamp);
    "#,
    },
+    Migration {
+        version: 3,
+        name: "add_owner_user_id",
+        sql: "ALTER TABLE visits ADD COLUMN owner_user_id INTEGER;",
+    },
 ];

 pub const SYSTEM_MIGRATIONS: &[Migration] = &[
@@ -384,4 +409,163 @@ pub const SYSTEM_MIGRATIONS: &[Migration] = &[
    CREATE INDEX IF NOT EXISTS idx_audit_action ON audit_events(action);
    "#,
    },
+    Migration {
+        version: 3,
+        name: "global_slugs_and_moderation",
+        sql: r#"
+    CREATE TABLE IF NOT EXISTS global_slugs (
+        slug TEXT PRIMARY KEY,
+        owner_user_id INTEGER NOT NULL,
+        target_type TEXT NOT NULL,
+        target_id TEXT NOT NULL,
+        created_at TEXT NOT NULL,
+        updated_at TEXT NOT NULL,
+        status TEXT NOT NULL,
+        deleted_at TEXT
+    );
+
+    CREATE INDEX IF NOT EXISTS idx_global_slugs_owner ON global_slugs(owner_user_id);
+    CREATE INDEX IF NOT EXISTS idx_global_slugs_status ON global_slugs(status);
+    CREATE INDEX IF NOT EXISTS idx_global_slugs_target ON global_slugs(target_type, target_id);
+
+    CREATE TABLE IF NOT EXISTS moderation_events (
+        id TEXT PRIMARY KEY,
+        timestamp TEXT NOT NULL,
+        admin_username TEXT NOT NULL,
+        target_user_id INTEGER NOT NULL,
+        target_username TEXT,
+        resource_type TEXT NOT NULL,
+        resource_identifier TEXT NOT NULL,
+        action TEXT NOT NULL,
+        severity TEXT NOT NULL,
+        reason TEXT NOT NULL
+    );
+
+    CREATE TABLE IF NOT EXISTS slug_history (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        slug TEXT NOT NULL,
+        old_owner_user_id INTEGER,
+        new_owner_user_id INTEGER,
+        action TEXT NOT NULL,
+        timestamp TEXT NOT NULL,
+        admin_username TEXT
+    );
+
+    CREATE TABLE IF NOT EXISTS reserved_slugs (
+        slug TEXT PRIMARY KEY,
+        reason TEXT
+    );
+
+    CREATE TABLE IF NOT EXISTS schema_version (
+        version INTEGER PRIMARY KEY,
+        applied_at TEXT NOT NULL
+    );
+
+    CREATE TABLE IF NOT EXISTS settings (
+        key TEXT PRIMARY KEY,
+        value TEXT NOT NULL
+    );
+
+    -- Seed defaults
+    INSERT OR IGNORE INTO schema_version (version, applied_at) VALUES (3, datetime('now'));
+    
+    INSERT OR IGNORE INTO settings (key, value) VALUES ('soft_delete_retention_days', '30');
+    INSERT OR IGNORE INTO settings (key, value) VALUES ('quota_reconcile_interval_hours', '24');
+    INSERT OR IGNORE INTO settings (key, value) VALUES ('allow_registration', 'false');
+    INSERT OR IGNORE INTO settings (key, value) VALUES ('maintenance_mode', 'false');
+
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('admin', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('login', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('logout', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('dashboard', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('api', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('docs', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('assets', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('static', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('favicon.ico', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('robots.txt', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('health', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('metrics', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('install', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('setup', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('support', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('help', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('security', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('abuse', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('billing', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('status', 'System route');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('legacy_admin', 'System reserved');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('administrator', 'System reserved');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('system', 'System reserved');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('root', 'System reserved');
+    INSERT OR IGNORE INTO reserved_slugs (slug, reason) VALUES ('www', 'System reserved');
+    "#,
+    },
+];
+
+pub const USERS_MIGRATIONS: &[Migration] = &[
+    Migration {
+        version: 1,
+        name: "initial_schema",
+        sql: r#"
+    CREATE TABLE IF NOT EXISTS users (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        username TEXT UNIQUE NOT NULL,
+        password_hash TEXT NOT NULL,
+        status TEXT NOT NULL DEFAULT 'active',
+        created_at TEXT NOT NULL,
+        last_login TEXT,
+        account_type TEXT DEFAULT 'standard',
+        organization_id INTEGER NULL,
+        metadata TEXT
+    );
+
+    CREATE TABLE IF NOT EXISTS quotas (
+        user_id INTEGER PRIMARY KEY,
+        max_urls INTEGER DEFAULT 100,
+        max_landings INTEGER DEFAULT 10,
+        max_api_tokens INTEGER DEFAULT 5,
+        max_storage_mb INTEGER DEFAULT 100,
+        current_urls INTEGER DEFAULT 0,
+        current_landings INTEGER DEFAULT 0,
+        current_api_tokens INTEGER DEFAULT 0,
+        current_storage_mb INTEGER DEFAULT 0,
+        FOREIGN KEY(user_id) REFERENCES users(id) ON DELETE CASCADE
+    );
+
+    CREATE TABLE IF NOT EXISTS api_tokens (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        user_id INTEGER NOT NULL,
+        token_hash TEXT NOT NULL,
+        created_at TEXT NOT NULL,
+        FOREIGN KEY(user_id) REFERENCES users(id) ON DELETE CASCADE
+    );
+
+    CREATE TABLE IF NOT EXISTS sessions (
+        id TEXT PRIMARY KEY,
+        user_id INTEGER NOT NULL,
+        expires_at TEXT NOT NULL,
+        created_at TEXT NOT NULL,
+        FOREIGN KEY(user_id) REFERENCES users(id) ON DELETE CASCADE
+    );
+
+    CREATE TABLE IF NOT EXISTS username_history (
+        id INTEGER PRIMARY KEY AUTOINCREMENT,
+        user_id INTEGER NOT NULL,
+        old_username TEXT NOT NULL,
+        new_username TEXT NOT NULL,
+        changed_at TEXT NOT NULL,
+        FOREIGN KEY(user_id) REFERENCES users(id) ON DELETE CASCADE
+    );
+    "#,
+    },
+    Migration {
+        version: 2,
+        name: "repair_admin_account_type",
+        sql: r#"
+    UPDATE users
+    SET account_type = 'admin'
+    WHERE username = 'admin' AND account_type = 'standard';
+    "#,
+    },
 ];
@@ -1,6 +1,7 @@
 use crate::config::Config;
 use crate::db::migrations::{
    run_migrations, ADMIN_MIGRATIONS, ANALYTICS_MIGRATIONS, CONTENT_MIGRATIONS, SYSTEM_MIGRATIONS,
+    USERS_MIGRATIONS,
 };
 use crate::db::sqlite::{enable_foreign_keys, enable_wal};
 use rusqlite::Connection;
@@ -15,6 +16,7 @@ pub mod migrations;
 pub mod preview;
 pub mod qr;
 pub mod sqlite;
+pub mod users;

 #[derive(Clone)]
 pub struct Db {
@@ -22,70 +24,108 @@ pub struct Db {
    pub content: Arc<Mutex<Connection>>,
    pub analytics: Arc<Mutex<Connection>>,
    pub system: Arc<Mutex<Connection>>,
+    pub users: Arc<Mutex<Connection>>,
+    pub data_dir: std::path::PathBuf,
 }

 impl Db {
    pub fn init(config: &Config) -> Result<Self, Box<dyn std::error::Error>> {
+        use chrono::Utc;
+        use tracing::info;
+
        // Ensure data directory exists
        if !config.data_dir.exists() {
            fs::create_dir_all(&config.data_dir)?;
        }

-        let admin_path = config.data_dir.join("admin.db");
-        let content_path = config.data_dir.join("content.db");
-        let analytics_path = config.data_dir.join("analytics.db");
-        let system_path = config.data_dir.join("system.db");
+        let admin_dir = config.data_dir.join("admin");
+        let users_dir = config.data_dir.join("users");
+        fs::create_dir_all(&admin_dir)?;
+        fs::create_dir_all(&users_dir)?;

-        use tracing::info;
+        // Automated Legacy Migration: check if legacy files are at the root
+        let legacy_admin_db = config.data_dir.join("admin.db");
+        let legacy_content_db = config.data_dir.join("content.db");
+        let legacy_analytics_db = config.data_dir.join("analytics.db");
+
+        // 1. If legacy admin.db exists at root, move admin/system DBs to config.data_dir/admin/
+        if legacy_admin_db.exists() {
+            tracing::warn!("LEGACY DETECTED: admin.db found at root. Moving administrative databases to multi-tenant admin/ subfolder...");
+            let files = vec![
+                "admin.db",
+                "admin.db-wal",
+                "admin.db-shm",
+                "system.db",
+                "system.db-wal",
+                "system.db-shm",
+            ];
+            for f in files {
+                let src = config.data_dir.join(f);
+                if src.exists() {
+                    let dst = admin_dir.join(f);
+                    let _ = fs::rename(&src, &dst);
+                }
+            }
+        }
+
+        // Pre-migration safety net: audit slug namespace for duplicates / format errors
+        match crate::db::users::audit_slug_namespace(config) {
+            Ok(report) => {
+                if !report.duplicates.is_empty() {
+                    tracing::error!(
+                        "Namespace conflicts detected before database migration: {:?}",
+                        report.duplicates
+                    );
+                    return Err(format!(
+                        "Database upgrade aborted due to slug conflicts: {:?}",
+                        report.duplicates
+                    )
+                    .into());
+                }
+            }
+            Err(e) => {
+                tracing::warn!("Failed to audit slug namespace before migration: {}", e);
+            }
+        }
+
+        let admin_path = admin_dir.join("admin.db");
+        let system_path = admin_dir.join("system.db");
+        let users_db_path = admin_dir.join("users.db");

        info!("Opening admin.db");
        let mut admin_conn = Connection::open(admin_path)?;
-        info!("Opening content.db");
-        let mut content_conn = Connection::open(content_path)?;
-        info!("Opening analytics.db");
-        let mut analytics_conn = Connection::open(analytics_path)?;
        info!("Opening system.db");
        let mut system_conn = Connection::open(system_path)?;
+        info!("Opening users.db");
+        let mut users_conn = Connection::open(users_db_path)?;

-        // Enable WAL mode for better concurrency and write performance
-        info!(database = "admin", "Enabling WAL mode on admin.db");
        enable_wal(&admin_conn, "admin")?;
-        info!(database = "content", "Enabling WAL mode on content.db");
-        enable_wal(&content_conn, "content")?;
-        info!(database = "analytics", "Enabling WAL mode on analytics.db");
-        enable_wal(&analytics_conn, "analytics")?;
-        info!(database = "system", "Enabling WAL mode on system.db");
        enable_wal(&system_conn, "system")?;
+        enable_wal(&users_conn, "users")?;

-        // Enable foreign key support
-        info!(
-            database = "admin",
-            "Enabling foreign key enforcement on admin.db"
-        );
        enable_foreign_keys(&admin_conn, "admin")?;
-        info!(
-            database = "content",
-            "Enabling foreign key enforcement on content.db"
-        );
-        enable_foreign_keys(&content_conn, "content")?;
-        info!(
-            database = "analytics",
-            "Enabling foreign key enforcement on analytics.db"
-        );
-        enable_foreign_keys(&analytics_conn, "analytics")?;
-        info!(
-            database = "system",
-            "Enabling foreign key enforcement on system.db"
-        );
        enable_foreign_keys(&system_conn, "system")?;
+        enable_foreign_keys(&users_conn, "users")?;

-        // 1. Run migrations for system.db first, as it receives secondary audit records
+        // Run migrations for system.db first
        info!("Running system migrations");
        run_migrations(&mut system_conn, "system", SYSTEM_MIGRATIONS, None)?;
-
        let system_arc = Arc::new(Mutex::new(system_conn));

-        // 2. Run migrations for other databases with system.db logging
+        // Pre-migration detection of admin account repair
+        let repair_needed = {
+            let stmt = users_conn.prepare(
+                "SELECT EXISTS(SELECT 1 FROM users WHERE username = 'admin' AND account_type = 'standard');"
+            );
+            match stmt {
+                Ok(mut s) => s
+                    .query_row([], |row| row.get::<_, bool>(0))
+                    .unwrap_or(false),
+                Err(_) => false,
+            }
+        };
+
+        // Run migrations for admin.db and users.db
        info!("Running admin migrations");
        run_migrations(
            &mut admin_conn,
@@ -93,14 +133,131 @@ impl Db {
            ADMIN_MIGRATIONS,
            Some(&system_arc),
        )?;
-        info!("Running content migrations");
+        info!("Running users migrations");
+        run_migrations(
+            &mut users_conn,
+            "users",
+            USERS_MIGRATIONS,
+            Some(&system_arc),
+        )?;
+
+        // Post-migration: audit log if repaired
+        if repair_needed {
+            let admin_is_now_admin: bool = users_conn
+                .query_row(
+                    "SELECT EXISTS(SELECT 1 FROM users WHERE username = 'admin' AND account_type = 'admin');",
+                    [],
+                    |row| row.get(0),
+                )
+                .unwrap_or(false);
+
+            if admin_is_now_admin {
+                let system_conn = system_arc.lock().unwrap();
+                let _ = crate::db::audit_events::write_audit_event(
+                    &system_conn,
+                    "admin",
+                    "migration_repair",
+                    "users",
+                    "admin",
+                    Some("Repaired standard account type to admin"),
+                );
+            }
+        }
+
+        // Clean up expired sessions from users.db on startup
+        let now = Utc::now().to_rfc3339();
+        let _ = users_conn.execute("DELETE FROM sessions WHERE expires_at < ?1;", [now]);
+
+        // 2. If legacy content.db/analytics.db exists, move them to users/1/ (for legacy_admin)
+        let legacy_migration_needed = legacy_content_db.exists() || legacy_analytics_db.exists();
+
+        // Ensure legacy_admin (user ID 1) exists in users.db
+        let legacy_admin_id = 1i64;
+        let legacy_admin_exists: bool = users_conn
+            .query_row(
+                "SELECT EXISTS(SELECT 1 FROM users WHERE id = ?1);",
+                [legacy_admin_id],
+                |row| row.get(0),
+            )
+            .unwrap_or(false);
+
+        if !legacy_admin_exists {
+            // Get copied administrator password hash
+            let admin_password_hash: String = admin_conn
+                .query_row(
+                    "SELECT password_hash FROM users ORDER BY created_at ASC LIMIT 1;",
+                    [],
+                    |row| row.get(0),
+                )
+                .unwrap_or_else(|_| {
+                    // If admin_db is empty, hash a default password
+                    crate::auth::password::hash_password("legacy_admin_pass").unwrap_or_default()
+                });
+
+            let now = Utc::now().to_rfc3339();
+            users_conn.execute(
+                "INSERT INTO users (id, username, password_hash, status, created_at, account_type) 
+                 VALUES (?1, ?2, ?3, ?4, ?5, ?6);",
+                rusqlite::params![
+                    legacy_admin_id,
+                    "legacy_admin",
+                    admin_password_hash,
+                    "disabled",
+                    now,
+                    "system"
+                ],
+            )?;
+
+            // Seed quotas
+            users_conn.execute(
+                "INSERT INTO quotas (user_id) VALUES (?1);",
+                [legacy_admin_id],
+            )?;
+        }
+
+        let legacy_user_dir = users_dir.join(legacy_admin_id.to_string());
+        fs::create_dir_all(&legacy_user_dir)?;
+
+        if legacy_content_db.exists() || legacy_analytics_db.exists() {
+            tracing::warn!("LEGACY DETECTED: content/analytics databases found at root. Moving to multi-tenant user ID 1 directory...");
+            let content_files = vec!["content.db", "content.db-wal", "content.db-shm"];
+            for f in content_files {
+                let src = config.data_dir.join(f);
+                if src.exists() {
+                    let dst = legacy_user_dir.join(f);
+                    let _ = fs::rename(&src, &dst);
+                }
+            }
+            let analytics_files = vec!["analytics.db", "analytics.db-wal", "analytics.db-shm"];
+            for f in analytics_files {
+                let src = config.data_dir.join(f);
+                if src.exists() {
+                    let dst = legacy_user_dir.join(f);
+                    let _ = fs::rename(&src, &dst);
+                }
+            }
+        }
+
+        // Open the legacy_admin databases (user ID 1) as db.content and db.analytics
+        let content_path = legacy_user_dir.join("content.db");
+        let analytics_path = legacy_user_dir.join("analytics.db");
+
+        let mut content_conn = Connection::open(content_path)?;
+        let mut analytics_conn = Connection::open(analytics_path)?;
+
+        enable_wal(&content_conn, "content")?;
+        enable_wal(&analytics_conn, "analytics")?;
+
+        enable_foreign_keys(&content_conn, "content")?;
+        enable_foreign_keys(&analytics_conn, "analytics")?;
+
+        // Run migrations for content.db and analytics.db
        run_migrations(
            &mut content_conn,
            "content",
            CONTENT_MIGRATIONS,
            Some(&system_arc),
        )?;
-        info!("Running analytics migrations");
        run_migrations(
            &mut analytics_conn,
            "analytics",
@@ -108,26 +265,256 @@ impl Db {
            Some(&system_arc),
        )?;

-        Ok(Self {
+        // If we just migrated legacy content, populate the global_slugs table in system.db
+        if legacy_migration_needed {
+            info!("Populating global slug index with legacy content...");
+            let mut sys_lock = system_arc.lock().unwrap();
+            let tx = sys_lock.transaction()?;
+
+            // Extract urls from content.db and insert into global_slugs
+            {
+                let mut stmt =
+                    content_conn.prepare("SELECT code, id, created_at, status FROM urls;")?;
+                let mut rows = stmt.query([])?;
+                while let Some(row) = rows.next()? {
+                    let slug: String = row.get(0)?;
+                    let target_id: String = row.get(1)?;
+                    let created_at: String = row.get(2)?;
+                    let status: String = row.get(3)?;
+                    let global_status = if status == "dead" {
+                        "disabled"
+                    } else {
+                        "active"
+                    };
+                    let now = Utc::now().to_rfc3339();
+
+                    let _ = tx.execute(
+                        "INSERT OR IGNORE INTO global_slugs (slug, owner_user_id, target_type, target_id, created_at, updated_at, status) 
+                         VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
+                        rusqlite::params![slug, legacy_admin_id, "url", target_id, created_at, now, global_status],
+                    );
+                }
+            }
+
+            // Extract landing pages from content.db and insert into global_slugs
+            {
+                let mut stmt = content_conn
+                    .prepare("SELECT code, id, created_at, state FROM landing_pages;")?;
+                let mut rows = stmt.query([])?;
+                while let Some(row) = rows.next()? {
+                    let slug: String = row.get(0)?;
+                    let target_id: String = row.get(1)?;
+                    let created_at: String = row.get(2)?;
+                    let state: String = row.get(3)?;
+                    let now = Utc::now().to_rfc3339();
+
+                    let status = if state == "published" {
+                        "active"
+                    } else {
+                        "disabled"
+                    };
+
+                    let _ = tx.execute(
+                        "INSERT OR IGNORE INTO global_slugs (slug, owner_user_id, target_type, target_id, created_at, updated_at, status) 
+                         VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
+                        rusqlite::params![slug, legacy_admin_id, "page", target_id, created_at, now, status],
+                    );
+                }
+            }
+
+            tx.commit()?;
+            info!("Global slug index populated successfully.");
+        }
+
+        let db = Self {
            admin: Arc::new(Mutex::new(admin_conn)),
            content: Arc::new(Mutex::new(content_conn)),
            analytics: Arc::new(Mutex::new(analytics_conn)),
            system: system_arc,
-        })
+            users: Arc::new(Mutex::new(users_conn)),
+            data_dir: config.data_dir.clone(),
+        };
+
+        let _ = db.reconcile_global_slugs(config);
+
+        // Post-init: Clean up stale reservations
+        {
+            let system_conn = db.system.lock().unwrap();
+            match crate::db::users::cleanup_stale_reservations(&system_conn, &config.data_dir) {
+                Ok(count) => {
+                    if count > 0 {
+                        tracing::info!("Cleaned up {} stale reserving slugs", count);
+                    }
+                }
+                Err(e) => {
+                    tracing::error!("Failed to clean up stale reservations: {}", e);
+                }
+            }
+        }
+
+        // Post-init: Verify global registry integrity
+        {
+            let system_conn = db.system.lock().unwrap();
+            let users_conn = db.users.lock().unwrap();
+            match crate::services::registry_validator::RegistryValidator::scan(
+                &system_conn,
+                &users_conn,
+                &config.data_dir,
+                None,
+            ) {
+                Ok(issues) => {
+                    for issue in issues {
+                        tracing::error!(
+                            "Global registry integrity issue: {:?} for slug {}",
+                            issue.issue_type,
+                            issue.slug
+                        );
+                    }
+                }
+                Err(e) => {
+                    tracing::error!("Failed to verify global registry integrity: {}", e);
+                }
+            }
+        }
+
+        Ok(db)
    }

    pub fn compact(&self) -> Result<(), rusqlite::Error> {
        let admin = self.admin.lock().unwrap();
-        admin.execute("VACUUM;", [])?;
+        let _ = admin.execute("VACUUM;", []);

        let content = self.content.lock().unwrap();
-        content.execute("VACUUM;", [])?;
+        let _ = content.execute("VACUUM;", []);

        let analytics = self.analytics.lock().unwrap();
-        analytics.execute("VACUUM;", [])?;
+        let _ = analytics.execute("VACUUM;", []);

        let system = self.system.lock().unwrap();
-        system.execute("VACUUM;", [])?;
+        let _ = system.execute("VACUUM;", []);
+
+        let users = self.users.lock().unwrap();
+        let _ = users.execute("VACUUM;", []);
+
+        Ok(())
+    }
+
+    pub fn init_user_databases(&self, user_id: i64) -> Result<(), Box<dyn std::error::Error>> {
+        let user_dir = self.data_dir.join("users").join(user_id.to_string());
+        fs::create_dir_all(&user_dir)?;
+
+        let content_path = user_dir.join("content.db");
+        let analytics_path = user_dir.join("analytics.db");
+        let profile_path = user_dir.join("profile.db");
+
+        let mut content_conn = Connection::open(content_path)?;
+        let mut analytics_conn = Connection::open(analytics_path)?;
+        let profile_conn = Connection::open(profile_path)?;
+
+        enable_wal(&content_conn, "content")?;
+        enable_wal(&analytics_conn, "analytics")?;
+        enable_wal(&profile_conn, "profile")?;
+
+        enable_foreign_keys(&content_conn, "content")?;
+        enable_foreign_keys(&analytics_conn, "analytics")?;
+        enable_foreign_keys(&profile_conn, "profile")?;
+
+        run_migrations(
+            &mut content_conn,
+            "content",
+            CONTENT_MIGRATIONS,
+            Some(&self.system),
+        )?;
+        run_migrations(
+            &mut analytics_conn,
+            "analytics",
+            ANALYTICS_MIGRATIONS,
+            Some(&self.system),
+        )?;
+
+        profile_conn.execute_batch(
+            "CREATE TABLE IF NOT EXISTS settings (
+                key TEXT PRIMARY KEY,
+                value TEXT NOT NULL
+            );",
+        )?;
+
+        Ok(())
+    }
+
+    pub fn reconcile_global_slugs(
+        &self,
+        config: &Config,
+    ) -> Result<(), Box<dyn std::error::Error>> {
+        use chrono::Utc;
+
+        let system_conn = self.system.lock().unwrap();
+        let users_conn = self.users.lock().unwrap();
+
+        // Get all user IDs
+        let mut stmt = users_conn.prepare("SELECT id FROM users;")?;
+        let mut rows = stmt.query([])?;
+        let mut user_ids = vec![1i64]; // Start with legacy admin
+        while let Some(row) = rows.next()? {
+            user_ids.push(row.get(0)?);
+        }
+        drop(rows);
+        drop(stmt);
+
+        for user_id in user_ids {
+            let user_dir = config.data_dir.join("users").join(user_id.to_string());
+            let content_path = user_dir.join("content.db");
+
+            if content_path.exists() {
+                let content_conn = Connection::open(&content_path)?;
+
+                // Sync URLs
+                let mut stmt =
+                    content_conn.prepare("SELECT code, id, created_at, status FROM urls;")?;
+                let mut rows = stmt.query([])?;
+                while let Some(row) = rows.next()? {
+                    let code: String = row.get(0)?;
+                    let target_id: String = row.get(1)?;
+                    let created_at: String = row.get(2)?;
+                    let status: String = row.get(3)?;
+                    let global_status = if status == "dead" {
+                        "disabled"
+                    } else {
+                        "active"
+                    };
+                    let now = Utc::now().to_rfc3339();
+
+                    let _ = system_conn.execute(
+                        "INSERT OR IGNORE INTO global_slugs (slug, owner_user_id, target_type, target_id, created_at, updated_at, status) 
+                         VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
+                        rusqlite::params![code, user_id, "url", target_id, created_at, now, global_status],
+                    );
+                }
+
+                // Sync Landing Pages
+                let mut stmt = content_conn
+                    .prepare("SELECT code, id, created_at, state FROM landing_pages;")?;
+                let mut rows = stmt.query([])?;
+                while let Some(row) = rows.next()? {
+                    let code: String = row.get(0)?;
+                    let target_id: String = row.get(1)?;
+                    let created_at: String = row.get(2)?;
+                    let state: String = row.get(3)?;
+                    let global_status = if state == "published" {
+                        "active"
+                    } else {
+                        "disabled"
+                    };
+                    let now = Utc::now().to_rfc3339();
+
+                    let _ = system_conn.execute(
+                        "INSERT OR IGNORE INTO global_slugs (slug, owner_user_id, target_type, target_id, created_at, updated_at, status) 
+                         VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
+                        rusqlite::params![code, user_id, "page", target_id, created_at, now, global_status],
+                    );
+                }
+            }
+        }

        Ok(())
    }
@@ -0,0 +1,869 @@
+use crate::models::{TenantUser, UserApiToken, UserQuotas, UserSession};
+use chrono::Utc;
+use rusqlite::{params, Connection, OptionalExtension};
+
+// --- User Operations ---
+
+pub fn is_reserved_username(username: &str) -> bool {
+    let u = username.trim().to_lowercase();
+    u == "admin" || u == "legacy_admin" || u == "administrator" || u == "system" || u == "root"
+}
+
+pub fn create_admin_user(
+    conn: &Connection,
+    username: &str,
+    password_hash: &str,
+) -> rusqlite::Result<TenantUser> {
+    let created_at = Utc::now().to_rfc3339();
+    let status = "active";
+    let account_type = "admin";
+
+    conn.execute(
+        "INSERT INTO users (username, password_hash, status, created_at, account_type, metadata) 
+         VALUES (?1, ?2, ?3, ?4, ?5, NULL);",
+        params![username, password_hash, status, created_at, account_type],
+    )?;
+
+    let id = conn.last_insert_rowid();
+
+    // Seed default quotas
+    conn.execute("INSERT INTO quotas (user_id) VALUES (?1);", params![id])?;
+
+    Ok(TenantUser {
+        id,
+        username: username.to_string(),
+        password_hash: password_hash.to_string(),
+        status: status.to_string(),
+        created_at,
+        last_login: None,
+        account_type: account_type.to_string(),
+        organization_id: None,
+        metadata: None,
+    })
+}
+
+pub fn create_user(
+    conn: &Connection,
+    username: &str,
+    password_hash: &str,
+    account_type: &str,
+    metadata: Option<&str>,
+) -> rusqlite::Result<TenantUser> {
+    if is_reserved_username(username) {
+        return Err(rusqlite::Error::SqliteFailure(
+            rusqlite::ffi::Error::new(rusqlite::ffi::SQLITE_CONSTRAINT),
+            Some("Username is reserved".to_string()),
+        ));
+    }
+
+    let created_at = Utc::now().to_rfc3339();
+    let status = "active";
+
+    conn.execute(
+        "INSERT INTO users (username, password_hash, status, created_at, account_type, metadata) 
+         VALUES (?1, ?2, ?3, ?4, ?5, ?6);",
+        params![
+            username,
+            password_hash,
+            status,
+            created_at,
+            account_type,
+            metadata
+        ],
+    )?;
+
+    let id = conn.last_insert_rowid();
+
+    // Seed default quotas
+    conn.execute("INSERT INTO quotas (user_id) VALUES (?1);", params![id])?;
+
+    Ok(TenantUser {
+        id,
+        username: username.to_string(),
+        password_hash: password_hash.to_string(),
+        status: status.to_string(),
+        created_at,
+        last_login: None,
+        account_type: account_type.to_string(),
+        organization_id: None,
+        metadata: metadata.map(|s| s.to_string()),
+    })
+}
+
+pub fn get_user_by_id(conn: &Connection, id: i64) -> rusqlite::Result<Option<TenantUser>> {
+    conn.query_row(
+        "SELECT id, username, password_hash, status, created_at, last_login, account_type, organization_id, metadata 
+         FROM users WHERE id = ?1;",
+        params![id],
+        |row| {
+            Ok(TenantUser {
+                id: row.get(0)?,
+                username: row.get(1)?,
+                password_hash: row.get(2)?,
+                status: row.get(3)?,
+                created_at: row.get(4)?,
+                last_login: row.get(5)?,
+                account_type: row.get(6)?,
+                organization_id: row.get(7)?,
+                metadata: row.get(8)?,
+            })
+        },
+    )
+    .optional()
+}
+
+pub fn get_user_by_username(
+    conn: &Connection,
+    username: &str,
+) -> rusqlite::Result<Option<TenantUser>> {
+    conn.query_row(
+        "SELECT id, username, password_hash, status, created_at, last_login, account_type, organization_id, metadata 
+         FROM users WHERE username = ?1;",
+        params![username],
+        |row| {
+            Ok(TenantUser {
+                id: row.get(0)?,
+                username: row.get(1)?,
+                password_hash: row.get(2)?,
+                status: row.get(3)?,
+                created_at: row.get(4)?,
+                last_login: row.get(5)?,
+                account_type: row.get(6)?,
+                organization_id: row.get(7)?,
+                metadata: row.get(8)?,
+            })
+        },
+    )
+    .optional()
+}
+
+pub fn delete_user(conn: &Connection, id: i64) -> rusqlite::Result<()> {
+    conn.execute("DELETE FROM users WHERE id = ?1;", params![id])?;
+    Ok(())
+}
+
+pub fn update_user_status(conn: &Connection, id: i64, status: &str) -> rusqlite::Result<()> {
+    conn.execute(
+        "UPDATE users SET status = ?1 WHERE id = ?2;",
+        params![status, id],
+    )?;
+    Ok(())
+}
+
+pub fn update_user_account_type(
+    conn: &Connection,
+    id: i64,
+    account_type: &str,
+) -> rusqlite::Result<()> {
+    conn.execute(
+        "UPDATE users SET account_type = ?1 WHERE id = ?2;",
+        params![account_type, id],
+    )?;
+    Ok(())
+}
+
+pub fn reset_user_password(
+    conn: &Connection,
+    id: i64,
+    new_password_hash: &str,
+) -> rusqlite::Result<()> {
+    conn.execute(
+        "UPDATE users SET password_hash = ?1 WHERE id = ?2;",
+        params![new_password_hash, id],
+    )?;
+    Ok(())
+}
+
+pub fn update_user_last_login(conn: &Connection, id: i64) -> rusqlite::Result<()> {
+    let now = Utc::now().to_rfc3339();
+    conn.execute(
+        "UPDATE users SET last_login = ?1 WHERE id = ?2;",
+        params![now, id],
+    )?;
+    Ok(())
+}
+
+pub fn list_users(conn: &Connection) -> rusqlite::Result<Vec<TenantUser>> {
+    let mut stmt = conn.prepare(
+        "SELECT id, username, password_hash, status, created_at, last_login, account_type, organization_id, metadata 
+         FROM users ORDER BY username ASC;",
+    )?;
+    let rows = stmt.query_map([], |row| {
+        Ok(TenantUser {
+            id: row.get(0)?,
+            username: row.get(1)?,
+            password_hash: row.get(2)?,
+            status: row.get(3)?,
+            created_at: row.get(4)?,
+            last_login: row.get(5)?,
+            account_type: row.get(6)?,
+            organization_id: row.get(7)?,
+            metadata: row.get(8)?,
+        })
+    })?;
+
+    let mut users = Vec::new();
+    for u in rows {
+        users.push(u?);
+    }
+    Ok(users)
+}
+
+pub fn log_username_change(
+    conn: &Connection,
+    user_id: i64,
+    old_username: &str,
+    new_username: &str,
+) -> rusqlite::Result<()> {
+    let now = Utc::now().to_rfc3339();
+    conn.execute(
+        "INSERT INTO username_history (user_id, old_username, new_username, changed_at) VALUES (?1, ?2, ?3, ?4);",
+        params![user_id, old_username, new_username, now],
+    )?;
+    conn.execute(
+        "UPDATE users SET username = ?1 WHERE id = ?2;",
+        params![new_username, user_id],
+    )?;
+    Ok(())
+}
+
+// --- Session Operations ---
+
+pub fn create_user_session(
+    conn: &Connection,
+    session_id: &str,
+    user_id: i64,
+    expires_at_rfc3339: &str,
+) -> rusqlite::Result<UserSession> {
+    let created_at = Utc::now().to_rfc3339();
+    conn.execute(
+        "INSERT INTO sessions (id, user_id, expires_at, created_at) VALUES (?1, ?2, ?3, ?4);",
+        params![session_id, user_id, expires_at_rfc3339, created_at],
+    )?;
+    Ok(UserSession {
+        id: session_id.to_string(),
+        user_id,
+        expires_at: expires_at_rfc3339.to_string(),
+        created_at,
+    })
+}
+
+pub fn get_user_session(
+    conn: &Connection,
+    session_id: &str,
+) -> rusqlite::Result<Option<UserSession>> {
+    conn.query_row(
+        "SELECT id, user_id, expires_at, created_at FROM sessions WHERE id = ?1;",
+        params![session_id],
+        |row| {
+            Ok(UserSession {
+                id: row.get(0)?,
+                user_id: row.get(1)?,
+                expires_at: row.get(2)?,
+                created_at: row.get(3)?,
+            })
+        },
+    )
+    .optional()
+}
+
+pub fn delete_user_session(conn: &Connection, session_id: &str) -> rusqlite::Result<()> {
+    conn.execute("DELETE FROM sessions WHERE id = ?1;", params![session_id])?;
+    Ok(())
+}
+
+pub fn cleanup_expired_user_sessions(conn: &Connection) -> rusqlite::Result<usize> {
+    let now = Utc::now().to_rfc3339();
+    let count = conn.execute("DELETE FROM sessions WHERE expires_at < ?1;", params![now])?;
+    Ok(count)
+}
+
+// --- Quota Operations ---
+
+pub fn get_user_quotas(conn: &Connection, user_id: i64) -> rusqlite::Result<Option<UserQuotas>> {
+    conn.query_row(
+        "SELECT user_id, max_urls, max_landings, max_api_tokens, max_storage_mb, 
+                current_urls, current_landings, current_api_tokens, current_storage_mb 
+         FROM quotas WHERE user_id = ?1;",
+        params![user_id],
+        |row| {
+            Ok(UserQuotas {
+                user_id: row.get(0)?,
+                max_urls: row.get(1)?,
+                max_landings: row.get(2)?,
+                max_api_tokens: row.get(3)?,
+                max_storage_mb: row.get(4)?,
+                current_urls: row.get(5)?,
+                current_landings: row.get(6)?,
+                current_api_tokens: row.get(7)?,
+                current_storage_mb: row.get(8)?,
+            })
+        },
+    )
+    .optional()
+}
+
+pub fn check_quota_limit(conn: &Connection, user_id: i64, field: &str) -> rusqlite::Result<bool> {
+    if let Some(quotas) = get_user_quotas(conn, user_id)? {
+        match field {
+            "urls" => Ok(quotas.current_urls < quotas.max_urls),
+            "landings" => Ok(quotas.current_landings < quotas.max_landings),
+            "api_tokens" => Ok(quotas.current_api_tokens < quotas.max_api_tokens),
+            _ => Ok(false),
+        }
+    } else {
+        Ok(false)
+    }
+}
+
+pub fn update_user_quotas(
+    conn: &Connection,
+    user_id: i64,
+    max_urls: i64,
+    max_landings: i64,
+    max_api_tokens: i64,
+    max_storage_mb: i64,
+) -> rusqlite::Result<()> {
+    conn.execute(
+        "UPDATE quotas SET max_urls = ?1, max_landings = ?2, max_api_tokens = ?3, max_storage_mb = ?4 
+         WHERE user_id = ?5;",
+        params![max_urls, max_landings, max_api_tokens, max_storage_mb, user_id],
+    )?;
+    Ok(())
+}
+
+pub fn increment_quota_counter(
+    conn: &Connection,
+    user_id: i64,
+    field: &str,
+) -> rusqlite::Result<()> {
+    let sql = match field {
+        "urls" => "UPDATE quotas SET current_urls = current_urls + 1 WHERE user_id = ?1;",
+        "landings" => {
+            "UPDATE quotas SET current_landings = current_landings + 1 WHERE user_id = ?1;"
+        }
+        "api_tokens" => {
+            "UPDATE quotas SET current_api_tokens = current_api_tokens + 1 WHERE user_id = ?1;"
+        }
+        _ => return Err(rusqlite::Error::InvalidQuery),
+    };
+    conn.execute(sql, params![user_id])?;
+    Ok(())
+}
+
+pub fn decrement_quota_counter(
+    conn: &Connection,
+    user_id: i64,
+    field: &str,
+) -> rusqlite::Result<()> {
+    let sql = match field {
+        "urls" => "UPDATE quotas SET current_urls = MAX(0, current_urls - 1) WHERE user_id = ?1;",
+        "landings" => "UPDATE quotas SET current_landings = MAX(0, current_landings - 1) WHERE user_id = ?1;",
+        "api_tokens" => "UPDATE quotas SET current_api_tokens = MAX(0, current_api_tokens - 1) WHERE user_id = ?1;",
+        _ => return Err(rusqlite::Error::InvalidQuery),
+    };
+    conn.execute(sql, params![user_id])?;
+    Ok(())
+}
+
+pub fn update_quota_storage(
+    conn: &Connection,
+    user_id: i64,
+    storage_mb: i64,
+) -> rusqlite::Result<()> {
+    conn.execute(
+        "UPDATE quotas SET current_storage_mb = ?1 WHERE user_id = ?2;",
+        params![storage_mb, user_id],
+    )?;
+    Ok(())
+}
+
+// --- API Token Operations ---
+
+pub fn create_user_api_token(
+    conn: &Connection,
+    user_id: i64,
+    token_hash: &str,
+) -> rusqlite::Result<UserApiToken> {
+    let created_at = Utc::now().to_rfc3339();
+    conn.execute(
+        "INSERT INTO api_tokens (user_id, token_hash, created_at) VALUES (?1, ?2, ?3);",
+        params![user_id, token_hash, created_at],
+    )?;
+    let id = conn.last_insert_rowid();
+
+    // Increment api token counter
+    let _ = increment_quota_counter(conn, user_id, "api_tokens");
+
+    Ok(UserApiToken {
+        id,
+        user_id,
+        token_hash: token_hash.to_string(),
+        created_at,
+    })
+}
+
+pub fn list_user_api_tokens(
+    conn: &Connection,
+    user_id: i64,
+) -> rusqlite::Result<Vec<UserApiToken>> {
+    let mut stmt = conn.prepare(
+        "SELECT id, user_id, token_hash, created_at FROM api_tokens WHERE user_id = ?1 ORDER BY id DESC;",
+    )?;
+    let rows = stmt.query_map(params![user_id], |row| {
+        Ok(UserApiToken {
+            id: row.get(0)?,
+            user_id: row.get(1)?,
+            token_hash: row.get(2)?,
+            created_at: row.get(3)?,
+        })
+    })?;
+
+    let mut tokens = Vec::new();
+    for t in rows {
+        tokens.push(t?);
+    }
+    Ok(tokens)
+}
+
+pub fn delete_user_api_token(conn: &Connection, id: i64, user_id: i64) -> rusqlite::Result<()> {
+    let deleted = conn.execute(
+        "DELETE FROM api_tokens WHERE id = ?1 AND user_id = ?2;",
+        params![id, user_id],
+    )?;
+    if deleted > 0 {
+        let _ = decrement_quota_counter(conn, user_id, "api_tokens");
+    }
+    Ok(())
+}
+
+// --- Global Slug & Quota Reconciliation Helpers ---
+
+pub fn is_slug_available(system_conn: &Connection, slug: &str) -> rusqlite::Result<bool> {
+    // 1. Check reserved list
+    let reserved: bool = system_conn
+        .query_row(
+            "SELECT EXISTS(SELECT 1 FROM reserved_slugs WHERE slug = ?1);",
+            [slug],
+            |row| row.get(0),
+        )
+        .unwrap_or(false);
+
+    if reserved {
+        return Ok(false);
+    }
+
+    // 2. Check global slugs
+    let exists: bool = system_conn
+        .query_row(
+            "SELECT EXISTS(SELECT 1 FROM global_slugs WHERE slug = ?1);",
+            [slug],
+            |row| row.get(0),
+        )
+        .unwrap_or(false);
+
+    Ok(!exists)
+}
+
+pub fn register_global_slug(
+    system_conn: &Connection,
+    slug: &str,
+    owner_user_id: i64,
+    target_type: &str,
+    target_id: &str,
+    status: &str,
+) -> rusqlite::Result<()> {
+    let now = Utc::now().to_rfc3339();
+    system_conn.execute(
+        "INSERT INTO global_slugs (slug, owner_user_id, target_type, target_id, created_at, updated_at, status)
+         VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
+        rusqlite::params![slug, owner_user_id, target_type, target_id, now, now, status],
+    )?;
+
+    // Insert history
+    system_conn.execute(
+        "INSERT INTO slug_history (slug, old_owner_user_id, new_owner_user_id, action, timestamp)
+         VALUES (?1, NULL, ?2, 'created', ?3);",
+        rusqlite::params![slug, owner_user_id, now],
+    )?;
+
+    Ok(())
+}
+
+pub fn release_global_slug(
+    system_conn: &Connection,
+    slug: &str,
+    owner_user_id: i64,
+) -> rusqlite::Result<()> {
+    let now = Utc::now().to_rfc3339();
+    system_conn.execute("DELETE FROM global_slugs WHERE slug = ?1;", [slug])?;
+
+    // Insert history
+    system_conn.execute(
+        "INSERT INTO slug_history (slug, old_owner_user_id, new_owner_user_id, action, timestamp)
+         VALUES (?1, ?2, NULL, 'released', ?3);",
+        rusqlite::params![slug, owner_user_id, now],
+    )?;
+
+    Ok(())
+}
+
+pub fn soft_delete_global_slug(
+    system_conn: &Connection,
+    slug: &str,
+    owner_user_id: i64,
+) -> rusqlite::Result<()> {
+    let now = Utc::now().to_rfc3339();
+    system_conn.execute(
+        "UPDATE global_slugs SET status = 'disabled', deleted_at = ?1 WHERE slug = ?2;",
+        rusqlite::params![now, slug],
+    )?;
+
+    // Insert history
+    system_conn.execute(
+        "INSERT INTO slug_history (slug, old_owner_user_id, new_owner_user_id, action, timestamp)
+         VALUES (?1, ?2, NULL, 'deleted', ?3);",
+        rusqlite::params![slug, owner_user_id, now],
+    )?;
+
+    Ok(())
+}
+
+#[derive(Clone, Debug, serde::Serialize, serde::Deserialize)]
+pub struct SlugAuditReport {
+    pub duplicates: Vec<String>,
+    pub invalid_entries: Vec<String>,
+    pub warnings: Vec<String>,
+}
+
+pub fn audit_slug_namespace(
+    config: &crate::config::Config,
+) -> Result<SlugAuditReport, Box<dyn std::error::Error>> {
+    use std::collections::HashMap;
+    let mut duplicates = Vec::new();
+    let mut invalid_entries = Vec::new();
+    let warnings = Vec::new();
+
+    let mut slug_owners: HashMap<String, Vec<i64>> = HashMap::new();
+
+    // 1. Scan legacy content.db if it exists
+    let legacy_content_path = config.data_dir.join("content.db");
+    if legacy_content_path.exists() {
+        if let Ok(conn) = Connection::open(&legacy_content_path) {
+            // URLs
+            if let Ok(mut stmt) = conn.prepare("SELECT code FROM urls;") {
+                if let Ok(mut rows) = stmt.query([]) {
+                    while let Some(row) = rows.next().unwrap_or(None) {
+                        if let Ok(code) = row.get::<_, String>(0) {
+                            slug_owners.entry(code).or_default().push(1); // 1 = legacy admin
+                        }
+                    }
+                }
+            }
+            // Landing Pages
+            if let Ok(mut stmt) = conn.prepare("SELECT code FROM landing_pages;") {
+                if let Ok(mut rows) = stmt.query([]) {
+                    while let Some(row) = rows.next().unwrap_or(None) {
+                        if let Ok(code) = row.get::<_, String>(0) {
+                            slug_owners.entry(code).or_default().push(1);
+                        }
+                    }
+                }
+            }
+        }
+    }
+
+    // 2. Scan all tenant databases in data_dir/users/<id>/content.db
+    let users_dir = config.data_dir.join("users");
+    if users_dir.exists() {
+        for entry in std::fs::read_dir(users_dir)? {
+            let entry = entry?;
+            let path = entry.path();
+            if path.is_dir() {
+                if let Some(name_str) = path.file_name().and_then(|n| n.to_str()) {
+                    if let Ok(user_id) = name_str.parse::<i64>() {
+                        let content_db_path = path.join("content.db");
+                        if content_db_path.exists() {
+                            if let Ok(conn) = Connection::open(&content_db_path) {
+                                // URLs
+                                if let Ok(mut stmt) = conn.prepare("SELECT code FROM urls;") {
+                                    if let Ok(mut rows) = stmt.query([]) {
+                                        while let Some(row) = rows.next().unwrap_or(None) {
+                                            if let Ok(code) = row.get::<_, String>(0) {
+                                                slug_owners.entry(code).or_default().push(user_id);
+                                            }
+                                        }
+                                    }
+                                }
+                                // Landing pages
+                                if let Ok(mut stmt) =
+                                    conn.prepare("SELECT code FROM landing_pages;")
+                                {
+                                    if let Ok(mut rows) = stmt.query([]) {
+                                        while let Some(row) = rows.next().unwrap_or(None) {
+                                            if let Ok(code) = row.get::<_, String>(0) {
+                                                slug_owners.entry(code).or_default().push(user_id);
+                                            }
+                                        }
+                                    }
+                                }
+                            }
+                        }
+                    }
+                }
+            }
+        }
+    }
+
+    // 3. Populate report
+    for (slug, owners) in slug_owners {
+        if owners.len() > 1 {
+            duplicates.push(format!(
+                "Slug '{}' is defined in multiple content databases by owners {:?}",
+                slug, owners
+            ));
+        }
+        // Validate slug format
+        let valid_url = crate::utils::validation::validate_redirect_code(&slug);
+        let valid_page = crate::utils::validation::validate_page_code(&slug);
+        if !valid_url && !valid_page {
+            invalid_entries.push(format!("Slug '{}' is format-invalid", slug));
+        }
+    }
+
+    Ok(SlugAuditReport {
+        duplicates,
+        invalid_entries,
+        warnings,
+    })
+}
+
+pub fn cleanup_stale_reservations(
+    system_conn: &Connection,
+    data_dir: &std::path::Path,
+) -> Result<usize, Box<dyn std::error::Error>> {
+    use chrono::{DateTime, Utc};
+    let mut cleaned_count = 0;
+
+    let mut stmt = system_conn.prepare(
+        "SELECT slug, owner_user_id, target_type, created_at FROM global_slugs WHERE status = 'reserving';"
+    )?;
+    let mut rows = stmt.query([])?;
+    let mut stale_slugs = Vec::new();
+
+    while let Some(row) = rows.next()? {
+        let slug: String = row.get(0)?;
+        let owner_user_id: i64 = row.get(1)?;
+        let target_type: String = row.get(2)?;
+        let created_at_str: String = row.get(3)?;
+
+        if let Ok(created_at) = DateTime::parse_from_rfc3339(&created_at_str) {
+            let age = Utc::now().signed_duration_since(created_at.with_timezone(&Utc));
+            if age > chrono::Duration::minutes(15) {
+                // Check if target record exists by looking up code = slug in owner's content.db
+                let content_db_path = if owner_user_id == 1 {
+                    let p1 = data_dir.join("users").join("1").join("content.db");
+                    if p1.exists() {
+                        p1
+                    } else {
+                        data_dir.join("content.db")
+                    }
+                } else {
+                    data_dir
+                        .join("users")
+                        .join(owner_user_id.to_string())
+                        .join("content.db")
+                };
+
+                let mut target_exists = false;
+                if content_db_path.exists() {
+                    if let Ok(conn) = Connection::open(&content_db_path) {
+                        if target_type == "url" {
+                            target_exists = conn
+                                .query_row(
+                                    "SELECT EXISTS(SELECT 1 FROM urls WHERE code = ?1);",
+                                    [&slug],
+                                    |r| r.get(0),
+                                )
+                                .unwrap_or(false);
+                        } else if target_type == "page" {
+                            target_exists = conn
+                                .query_row(
+                                    "SELECT EXISTS(SELECT 1 FROM landing_pages WHERE code = ?1);",
+                                    [&slug],
+                                    |r| r.get(0),
+                                )
+                                .unwrap_or(false);
+                        }
+                    }
+                }
+
+                if !target_exists {
+                    stale_slugs.push((slug, owner_user_id));
+                }
+            }
+        }
+    }
+
+    drop(rows);
+    drop(stmt);
+
+    for (slug, owner_user_id) in stale_slugs {
+        system_conn.execute("DELETE FROM global_slugs WHERE slug = ?1;", [&slug])?;
+        let now = Utc::now().to_rfc3339();
+        system_conn.execute(
+            "INSERT INTO slug_history (slug, old_owner_user_id, new_owner_user_id, action, timestamp)
+             VALUES (?1, ?2, NULL, 'released', ?3);",
+            rusqlite::params![slug, owner_user_id, now],
+        )?;
+        cleaned_count += 1;
+    }
+
+    Ok(cleaned_count)
+}
+
+pub fn register_restored_user_slugs(
+    system_conn: &Connection,
+    target_user_id: i64,
+    restored_content_db_path: &std::path::Path,
+) -> Result<(), Box<dyn std::error::Error>> {
+    let restored_content_conn = Connection::open(restored_content_db_path)?;
+
+    let mut urls = Vec::new();
+    let mut landing_pages = Vec::new();
+
+    // 1. Read URLs
+    {
+        let mut stmt =
+            restored_content_conn.prepare("SELECT code, id, created_at, status FROM urls;")?;
+        let mut rows = stmt.query([])?;
+        while let Some(row) = rows.next()? {
+            let code: String = row.get(0)?;
+            let id: String = row.get(1)?;
+            let created_at: String = row.get(2)?;
+            let status: String = row.get(3)?;
+            urls.push((code, id, created_at, status));
+        }
+    }
+
+    // 2. Read Landing Pages
+    {
+        let mut stmt = restored_content_conn
+            .prepare("SELECT code, id, created_at, state FROM landing_pages;")?;
+        let mut rows = stmt.query([])?;
+        while let Some(row) = rows.next()? {
+            let code: String = row.get(0)?;
+            let id: String = row.get(1)?;
+            let created_at: String = row.get(2)?;
+            let state: String = row.get(3)?;
+            landing_pages.push((code, id, created_at, state));
+        }
+    }
+
+    // 3. Check for collisions across all URLs and landing pages
+    let mut conflicting_slugs = Vec::new();
+    for (slug, _, _, _) in &urls {
+        let existing_owner: Option<i64> = system_conn
+            .query_row(
+                "SELECT owner_user_id FROM global_slugs WHERE slug = ?1;",
+                [slug],
+                |r| r.get(0),
+            )
+            .optional()?;
+
+        if let Some(owner) = existing_owner {
+            if owner != target_user_id {
+                conflicting_slugs.push(slug.clone());
+            }
+        }
+    }
+
+    for (slug, _, _, _) in &landing_pages {
+        let existing_owner: Option<i64> = system_conn
+            .query_row(
+                "SELECT owner_user_id FROM global_slugs WHERE slug = ?1;",
+                [slug],
+                |r| r.get(0),
+            )
+            .optional()?;
+
+        if let Some(owner) = existing_owner {
+            if owner != target_user_id {
+                conflicting_slugs.push(slug.clone());
+            }
+        }
+    }
+
+    if !conflicting_slugs.is_empty() {
+        return Err(format!(
+            "Restore failed. Conflicting slugs: {}",
+            conflicting_slugs.join(", ")
+        )
+        .into());
+    }
+
+    // 4. Perform registration
+    system_conn.execute(
+        "DELETE FROM global_slugs WHERE owner_user_id = ?1;",
+        [target_user_id],
+    )?;
+
+    for (slug, target_id, created_at, status) in urls {
+        let now = Utc::now().to_rfc3339();
+        let global_status = if status == "dead" {
+            "disabled"
+        } else {
+            "active"
+        };
+        system_conn.execute(
+            "INSERT OR REPLACE INTO global_slugs (slug, owner_user_id, target_type, target_id, created_at, updated_at, status) 
+             VALUES (?1, ?2, 'url', ?3, ?4, ?5, ?6);",
+            rusqlite::params![slug, target_user_id, target_id, created_at, now, global_status],
+        )?;
+    }
+
+    for (slug, target_id, created_at, state) in landing_pages {
+        let now = Utc::now().to_rfc3339();
+        let status = if state == "published" {
+            "active"
+        } else {
+            "disabled"
+        };
+        system_conn.execute(
+            "INSERT OR REPLACE INTO global_slugs (slug, owner_user_id, target_type, target_id, created_at, updated_at, status) 
+             VALUES (?1, ?2, 'page', ?3, ?4, ?5, ?6);",
+            rusqlite::params![slug, target_user_id, target_id, created_at, now, status],
+        )?;
+    }
+
+    Ok(())
+}
+
+pub fn reconcile_user_quotas(
+    users_conn: &Connection,
+    user_id: i64,
+    content_conn: &Connection,
+) -> rusqlite::Result<()> {
+    let urls_count: i64 = content_conn
+        .query_row("SELECT COUNT(*) FROM urls;", [], |row| row.get(0))
+        .unwrap_or(0);
+
+    let landings_count: i64 = content_conn
+        .query_row("SELECT COUNT(*) FROM landing_pages;", [], |row| row.get(0))
+        .unwrap_or(0);
+
+    let api_tokens_count: i64 = users_conn
+        .query_row(
+            "SELECT COUNT(*) FROM api_tokens WHERE user_id = ?1;",
+            [user_id],
+            |row| row.get(0),
+        )
+        .unwrap_or(0);
+
+    users_conn.execute(
+        "UPDATE quotas SET current_urls = ?1, current_landings = ?2, current_api_tokens = ?3 WHERE user_id = ?4;",
+        rusqlite::params![urls_count, landings_count, api_tokens_count, user_id],
+    )?;
+
+    Ok(())
+}
@@ -5,28 +5,58 @@ use super::{log_job_end, log_job_start};
 use crate::analytics::aggregate_day;
 use crate::db::Db;

-pub async fn run_aggregator(db: Db, interval_mins: u64) {
+pub async fn run_aggregator(
+    db: Db,
+    interval_mins: u64,
+    mut shutdown_rx: tokio::sync::watch::Receiver<bool>,
+) {
    loop {
-        tokio::time::sleep(Duration::from_secs(interval_mins * 60)).await;
+        tokio::select! {
+            _ = tokio::time::sleep(Duration::from_secs(interval_mins * 60)) => {}
+            _ = shutdown_rx.changed() => {
+                info!("Analytics aggregator shutting down...");
+                break;
+            }
+        }
        info!("Running background analytics aggregator...");

+        let user_ids: Vec<i64> = {
+            let conn = db.users.lock().unwrap();
+            let mut stmt = match conn.prepare("SELECT id FROM users;") {
+                Ok(s) => s,
+                Err(_) => continue,
+            };
+            let rows = match stmt.query_map([], |row| row.get(0)) {
+                Ok(r) => r,
+                Err(_) => continue,
+            };
+            rows.filter_map(|r| r.ok()).collect()
+        };
+
        let job_id = log_job_start(&db.system, "analytics_aggregator");
-        match perform_aggregation(&db).await {
-            Ok(_) => log_job_end(&db.system, &job_id, "success", None),
-            Err(e) => {
-                let err_str = e.to_string();
-                error!("Error performing aggregation: {}", err_str);
-                log_job_end(&db.system, &job_id, "failed", Some(&err_str));
+        let mut failed = false;
+        let mut err_msg = None;
+
+        for user_id in user_ids {
+            if let Err(e) = perform_aggregation(&db, user_id).await {
+                failed = true;
+                err_msg = Some(e.to_string());
            }
        }
+
+        if failed {
+            let err_str = err_msg.unwrap_or_else(|| "Unknown error".to_string());
+            error!("Error performing aggregation: {}", err_str);
+            log_job_end(&db.system, &job_id, "failed", Some(&err_str));
+        } else {
+            log_job_end(&db.system, &job_id, "success", None);
+        }
    }
 }

-pub async fn perform_aggregation(db: &Db) -> Result<(), Box<dyn std::error::Error>> {
-    let date_range = {
-        let conn = db.analytics.lock().unwrap();
-        crate::db::analytics::get_visits_date_range(&conn)?
-    };
+pub async fn perform_aggregation(db: &Db, user_id: i64) -> Result<(), Box<dyn std::error::Error>> {
+    let mut conn = super::open_user_analytics_conn(db, user_id)?;
+    let date_range = crate::db::analytics::get_visits_date_range(&conn)?;

    if let Some((min_date, max_date)) = date_range {
        let min = chrono::NaiveDate::parse_from_str(&min_date, "%Y-%m-%d")?;
@@ -35,10 +65,7 @@ pub async fn perform_aggregation(db: &Db) -> Result<(), Box<dyn std::error::Erro
        let mut curr = min;
        while curr <= max {
            let date_str = curr.format("%Y-%m-%d").to_string();
-            {
-                let mut conn = db.analytics.lock().unwrap();
-                aggregate_day(&mut conn, &date_str)?;
-            }
+            aggregate_day(&mut conn, &date_str)?;
            if curr == max {
                break;
            }
@@ -4,7 +4,11 @@ use crate::db::Db;
 use std::time::Duration;
 use tracing::{error, info};

-pub async fn run_backup_scheduler(db: Db, config: Config) {
+pub async fn run_backup_scheduler(
+    db: Db,
+    config: Config,
+    mut shutdown_rx: tokio::sync::watch::Receiver<bool>,
+) {
    if !config.backup_enabled {
        info!("Background backup scheduler is disabled.");
        return;
@@ -16,7 +20,13 @@ pub async fn run_backup_scheduler(db: Db, config: Config) {
    );
    loop {
        // Run backup every configured interval
-        tokio::time::sleep(Duration::from_secs(config.backup_interval_mins * 60)).await;
+        tokio::select! {
+            _ = tokio::time::sleep(Duration::from_secs(config.backup_interval_mins * 60)) => {}
+            _ = shutdown_rx.changed() => {
+                info!("Backup scheduler shutting down...");
+                break;
+            }
+        }
        info!("Running background database backup...");

        let job_id = log_job_start(&db.system, "database_backup");
@@ -51,6 +61,36 @@ pub async fn perform_backup(
        std::fs::create_dir_all(&out_dir)?;
    }

+    // Force checkpoint on all databases to flush WAL contents to the main DB files
+    if let Ok(conn) = db.admin.lock() {
+        let _ = conn.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+    }
+    if let Ok(conn) = db.content.lock() {
+        let _ = conn.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+    }
+    if let Ok(conn) = db.analytics.lock() {
+        let _ = conn.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+    }
+    if let Ok(conn) = db.system.lock() {
+        let _ = conn.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+    }
+    if let Ok(conn) = db.users.lock() {
+        let _ = conn.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+        if let Ok(mut stmt) = conn.prepare("SELECT id FROM users;") {
+            if let Ok(rows) = stmt.query_map([], |row| row.get::<_, i64>(0)) {
+                let user_ids: Vec<i64> = rows.filter_map(|r| r.ok()).collect();
+                for user_id in user_ids {
+                    if let Ok(u_conn) = crate::jobs::open_user_content_conn(db, user_id) {
+                        let _ = u_conn.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+                    }
+                    if let Ok(u_conn) = crate::jobs::open_user_analytics_conn(db, user_id) {
+                        let _ = u_conn.execute("PRAGMA wal_checkpoint(TRUNCATE);", []);
+                    }
+                }
+            }
+        }
+    }
+
    let date_str = Utc::now().format("%Y-%m-%d-%H%M%S").to_string();
    let tar_name = format!("{}-bzod-backup.tar.gz", date_str);
    let tar_path = out_dir.join(tar_name);
@@ -59,12 +99,14 @@ pub async fn perform_backup(
    let enc = GzEncoder::new(file, Compression::default());
    let mut tar = Builder::new(enc);

-    let files = vec!["admin.db", "content.db", "analytics.db", "system.db"];
-    for f in files {
-        let db_file = config.data_dir.join(f);
-        if db_file.exists() {
-            tar.append_path_with_name(&db_file, f)?;
-        }
+    let admin_dir = config.data_dir.join("admin");
+    if admin_dir.exists() {
+        tar.append_dir_all("admin", &admin_dir)?;
+    }
+
+    let users_dir = config.data_dir.join("users");
+    if users_dir.exists() {
+        tar.append_dir_all("users", &users_dir)?;
    }

    tar.into_inner()?.finish()?;
@@ -1,22 +1,81 @@
 use crate::db::Db;
 use std::time::Duration;
-use tracing::info;
+use tracing::{error, info, warn};

 /// Background job that marks expired URLs.
 ///
 /// Runs every 60 seconds. Any URL with `expires_at < NOW()` and `expired = 0`
 /// gets flipped to `expired = 1`.
-pub async fn run_expiry_checker(db: Db) {
+///
+/// Correctness note: the redirect handler treats wall-clock `expires_at` as
+/// authoritative and returns 410 without depending on this sweeper. The sweeper
+/// is maintenance (persist `expired=1`) and must remain idempotent.
+pub async fn run_expiry_checker(db: Db, mut shutdown_rx: tokio::sync::watch::Receiver<bool>) {
    loop {
-        tokio::time::sleep(Duration::from_secs(60)).await;
+        tokio::select! {
+            _ = tokio::time::sleep(Duration::from_secs(60)) => {}
+            _ = shutdown_rx.changed() => {
+                info!("Expiry checker shutting down...");
+                break;
+            }
+        }

-        let count = {
-            let conn = db.content.lock().unwrap();
-            crate::db::content::expire_urls(&conn).unwrap_or(0)
+        let user_ids: Vec<i64> = {
+            let conn = match db.users.lock() {
+                Ok(c) => c,
+                Err(e) => {
+                    error!(error = %e, "expiry job: users_db mutex poisoned");
+                    continue;
+                }
+            };
+            let mut stmt = match conn.prepare("SELECT id FROM users;") {
+                Ok(s) => s,
+                Err(e) => {
+                    error!(error = %e, "expiry job: failed to list users");
+                    continue;
+                }
+            };
+            let rows = match stmt.query_map([], |row| row.get(0)) {
+                Ok(r) => r,
+                Err(e) => {
+                    error!(error = %e, "expiry job: failed to map user ids");
+                    continue;
+                }
+            };
+            rows.filter_map(|r| r.ok()).collect()
        };

-        if count > 0 {
-            info!(expired_count = count, "Expired URLs marked");
+        let mut total_expired = 0;
+        for user_id in user_ids {
+            match super::open_user_content_conn(&db, user_id) {
+                Ok(conn) => match crate::db::content::expire_urls(&conn) {
+                    Ok(count) => total_expired += count,
+                    Err(e) => {
+                        warn!(
+                            owner_user_id = user_id,
+                            error = %e,
+                            "expiry job: expire_urls failed"
+                        );
+                    }
+                },
+                Err(e) => {
+                    // Missing content.db for a user is common; only log open errors that are unexpected.
+                    if !matches!(e, rusqlite::Error::SqliteFailure(_, _)) {
+                        warn!(
+                            owner_user_id = user_id,
+                            error = %e,
+                            "expiry job: could not open content.db"
+                        );
+                    }
+                }
+            }
+        }
+
+        if total_expired > 0 {
+            info!(
+                expired_count = total_expired,
+                "Expired URLs marked across users"
+            );
        }
    }
 }
@@ -8,7 +8,11 @@ use uuid::Uuid;
 use super::{log_job_end, log_job_start};
 use crate::db::Db;

-pub async fn run_link_checker(db: Db, interval_mins: u64) {
+pub async fn run_link_checker(
+    db: Db,
+    interval_mins: u64,
+    mut shutdown_rx: tokio::sync::watch::Receiver<bool>,
+) {
    let client = Client::builder()
        .timeout(Duration::from_secs(10))
        .user_agent("bzod-link-checker/0.1")
@@ -18,7 +22,14 @@ pub async fn run_link_checker(db: Db, interval_mins: u64) {

    loop {
        // Sleep first to give server time to start up
-        tokio::time::sleep(Duration::from_secs(interval_mins * 60)).await;
+        tokio::select! {
+            _ = tokio::time::sleep(Duration::from_secs(interval_mins * 60)) => {}
+            _ = shutdown_rx.changed() => {
+                info!("Link checker shutting down...");
+                break;
+            }
+        }
+
        info!("Running background link health check...");

        let job_id = log_job_start(&db.system, "link_checker");
@@ -37,40 +48,50 @@ pub async fn perform_link_check(
    db: &Db,
    client: &Client,
 ) -> Result<(), Box<dyn std::error::Error>> {
-    let urls = {
-        let conn = db.content.lock().unwrap();
-        crate::db::content::list_urls_for_health_check(&conn)?
+    let user_ids: Vec<i64> = {
+        let conn = db.users.lock().unwrap();
+        let mut stmt = conn.prepare("SELECT id FROM users;")?;
+        let rows = stmt.query_map([], |row| row.get(0))?;
+        rows.filter_map(|r| r.ok()).collect()
    };

-    for (id, dest) in urls {
-        let (status, detail_status, status_code, latency_ms, err_msg) =
-            check_url_health(client, &dest).await;
-        {
-            let conn = db.content.lock().unwrap();
-            crate::db::content::update_url_health_extended(
-                &conn,
-                &id,
-                &status,
-                &detail_status,
-                Some(latency_ms),
-            )?;
-        }
+    for user_id in user_ids {
+        let conn = match super::open_user_content_conn(db, user_id) {
+            Ok(c) => c,
+            Err(_) => continue,
+        };

-        // Log to system.db.health_checks
-        {
-            let conn = db.system.lock().unwrap();
-            let hc_id = Uuid::new_v4().to_string();
-            let now = Utc::now().to_rfc3339();
-            let is_healthy = if status == "healthy" { 1 } else { 0 };
-            let _ = conn.execute(
-                "INSERT INTO health_checks (id, object_type, object_id, checked_at, status_code, error_message, is_healthy) 
-                 VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
-                params![hc_id, "url", id, now, status_code, err_msg, is_healthy],
-            );
-        }
+        let urls = crate::db::content::list_urls_for_health_check(&conn)?;

-        // Rate limiting sleep between external requests
-        tokio::time::sleep(Duration::from_millis(200)).await;
+        for (id, dest) in urls {
+            let (status, detail_status, status_code, latency_ms, err_msg) =
+                check_url_health(client, &dest).await;
+            {
+                crate::db::content::update_url_health_extended(
+                    &conn,
+                    &id,
+                    &status,
+                    &detail_status,
+                    Some(latency_ms),
+                )?;
+            }
+
+            // Log to system.db.health_checks
+            {
+                let sys_conn = db.system.lock().unwrap();
+                let hc_id = Uuid::new_v4().to_string();
+                let now = Utc::now().to_rfc3339();
+                let is_healthy = if status == "healthy" { 1 } else { 0 };
+                let _ = sys_conn.execute(
+                    "INSERT INTO health_checks (id, object_type, object_id, checked_at, status_code, error_message, is_healthy) 
+                     VALUES (?1, ?2, ?3, ?4, ?5, ?6, ?7);",
+                    params![hc_id, "url", id, now, status_code, err_msg, is_healthy],
+                );
+            }
+
+            // Rate limiting sleep between external requests
+            tokio::time::sleep(Duration::from_millis(200)).await;
+        }
    }
    Ok(())
 }
@@ -1,3 +1,4 @@
+use crate::db::Db;
 use chrono::Utc;
 use rusqlite::{params, Connection};
 use std::sync::Mutex;
@@ -35,3 +36,38 @@ pub fn log_job_end(conn: &Mutex<Connection>, id: &str, status: &str, err_msg: Op
        );
    }
 }
+
+pub mod quota_reconcile;
+pub use quota_reconcile::run_quota_reconciliation;
+
+// --- Database Connection Helpers for User-specific Databases ---
+
+pub fn open_user_content_conn(
+    db: &Db,
+    user_id: i64,
+) -> Result<rusqlite::Connection, rusqlite::Error> {
+    let db_path = db
+        .data_dir
+        .join("users")
+        .join(user_id.to_string())
+        .join("content.db");
+    let conn = rusqlite::Connection::open(db_path)?;
+    crate::db::sqlite::enable_wal(&conn, "content")?;
+    crate::db::sqlite::enable_foreign_keys(&conn, "content")?;
+    Ok(conn)
+}
+
+pub fn open_user_analytics_conn(
+    db: &Db,
+    user_id: i64,
+) -> Result<rusqlite::Connection, rusqlite::Error> {
+    let db_path = db
+        .data_dir
+        .join("users")
+        .join(user_id.to_string())
+        .join("analytics.db");
+    let conn = rusqlite::Connection::open(db_path)?;
+    crate::db::sqlite::enable_wal(&conn, "analytics")?;
+    crate::db::sqlite::enable_foreign_keys(&conn, "analytics")?;
+    Ok(conn)
+}
@@ -0,0 +1,52 @@
+use crate::db::Db;
+use std::time::Duration;
+use tracing::{error, info};
+
+pub async fn run_quota_reconciliation(
+    db: Db,
+    interval_hours: u64,
+    mut shutdown_rx: tokio::sync::watch::Receiver<bool>,
+) {
+    loop {
+        // Sleep first
+        tokio::select! {
+            _ = tokio::time::sleep(Duration::from_secs(interval_hours * 3600)) => {}
+            _ = shutdown_rx.changed() => {
+                info!("Quota reconciliation shutting down...");
+                break;
+            }
+        }
+        info!("Running background quota reconciliation...");
+
+        let user_ids: Vec<i64> = {
+            let conn = db.users.lock().unwrap();
+            let mut stmt = match conn.prepare("SELECT id FROM users;") {
+                Ok(s) => s,
+                Err(e) => {
+                    error!("Failed to prepare select user IDs: {:?}", e);
+                    continue;
+                }
+            };
+            let rows = match stmt.query_map([], |row| row.get(0)) {
+                Ok(r) => r,
+                Err(e) => {
+                    error!("Failed to query user IDs: {:?}", e);
+                    continue;
+                }
+            };
+            rows.filter_map(|r| r.ok()).collect()
+        };
+
+        let users_conn = db.users.lock().unwrap();
+        for user_id in user_ids {
+            if let Ok(content_conn) = super::open_user_content_conn(&db, user_id) {
+                if let Err(e) =
+                    crate::db::users::reconcile_user_quotas(&users_conn, user_id, &content_conn)
+                {
+                    error!("Failed to reconcile quotas for user {}: {:?}", user_id, e);
+                }
+            }
+        }
+        info!("Quota reconciliation finished.");
+    }
+}
@@ -4,7 +4,11 @@ use tracing::{error, info};
 use super::{log_job_end, log_job_start};
 use crate::db::Db;

-pub async fn run_retention_cleaner(db: Db, retention_days_opt: Option<i64>) {
+pub async fn run_retention_cleaner(
+    db: Db,
+    retention_days_opt: Option<i64>,
+    mut shutdown_rx: tokio::sync::watch::Receiver<bool>,
+) {
    let retention_days = match retention_days_opt {
        Some(days) => days,
        None => return,
@@ -12,21 +16,59 @@ pub async fn run_retention_cleaner(db: Db, retention_days_opt: Option<i64>) {

    loop {
        // Check once every 24 hours
-        tokio::time::sleep(Duration::from_secs(24 * 3600)).await;
+        tokio::select! {
+            _ = tokio::time::sleep(Duration::from_secs(24 * 3600)) => {}
+            _ = shutdown_rx.changed() => {
+                info!("Retention cleaner shutting down...");
+                break;
+            }
+        }
        info!("Running background data retention cleanup...");

+        let user_ids: Vec<i64> = {
+            let conn = db.users.lock().unwrap();
+            let mut stmt = match conn.prepare("SELECT id FROM users;") {
+                Ok(s) => s,
+                Err(_) => continue,
+            };
+            let rows = match stmt.query_map([], |row| row.get(0)) {
+                Ok(r) => r,
+                Err(_) => continue,
+            };
+            rows.filter_map(|r| r.ok()).collect()
+        };
+
        let job_id = log_job_start(&db.system, "retention_cleaner");
-        let conn = db.analytics.lock().unwrap();
-        match crate::db::analytics::retention_cleanup(&conn, retention_days) {
-            Ok(count) => {
-                info!("Cleaned up {} expired visits from database", count);
-                log_job_end(&db.system, &job_id, "success", None);
-            }
-            Err(e) => {
-                let err_str = e.to_string();
-                error!("Error running retention cleaner: {:?}", err_str);
-                log_job_end(&db.system, &job_id, "failed", Some(&err_str));
+        let mut total_cleaned = 0;
+        let mut failed = false;
+        let mut err_msg = None;
+
+        for user_id in user_ids {
+            match super::open_user_analytics_conn(&db, user_id) {
+                Ok(conn) => match crate::db::analytics::retention_cleanup(&conn, retention_days) {
+                    Ok(count) => total_cleaned += count,
+                    Err(e) => {
+                        failed = true;
+                        err_msg = Some(e.to_string());
+                    }
+                },
+                Err(e) => {
+                    failed = true;
+                    err_msg = Some(e.to_string());
+                }
            }
        }
+
+        if failed {
+            let err_str = err_msg.unwrap_or_else(|| "Unknown error".to_string());
+            error!("Error running retention cleaner: {:?}", err_str);
+            log_job_end(&db.system, &job_id, "failed", Some(&err_str));
+        } else {
+            info!(
+                "Cleaned up {} expired visits across all user databases",
+                total_cleaned
+            );
+            log_job_end(&db.system, &job_id, "success", None);
+        }
    }
 }
@@ -38,12 +38,77 @@ async fn main() -> Result<(), Box<dyn std::error::Error>> {
        Commands::Validate { data_dir } => {
            bzod::cli::validate::run(data_dir, config).await?;
        }
+        Commands::AuditDestinations { data_dir } => {
+            bzod::cli::audit_destinations::run(data_dir, config).await?;
+        }
        Commands::CreateAdmin { username, data_dir } => {
            bzod::cli::create_admin::run(username, data_dir, config).await?;
        }
        Commands::Doctor { data_dir } => {
            bzod::cli::doctor::run(data_dir, config).await?;
        }
+        Commands::Shorten {
+            target_url,
+            slug,
+            data_dir,
+        } => {
+            bzod::cli::shorten::run(target_url, slug, data_dir, config).await?;
+        }
+        Commands::Expand { code, data_dir } => {
+            bzod::cli::expand::run(code, data_dir, config).await?;
+        }
+        Commands::CreateUser {
+            username,
+            password,
+            data_dir,
+        } => {
+            bzod::cli::create_user::run(username, password, data_dir, config).await?;
+        }
+        Commands::DeleteUser {
+            user_id,
+            force,
+            data_dir,
+        } => {
+            bzod::cli::delete_user::run(user_id, force, data_dir, config).await?;
+        }
+        Commands::DisableUser { user_id, data_dir } => {
+            bzod::cli::disable_user::run(user_id, data_dir, config).await?;
+        }
+        Commands::EnableUser { user_id, data_dir } => {
+            bzod::cli::enable_user::run(user_id, data_dir, config).await?;
+        }
+        Commands::ResetPassword {
+            user_id,
+            password,
+            data_dir,
+        } => {
+            bzod::cli::reset_password::run(user_id, password, data_dir, config).await?;
+        }
+        Commands::ListUsers { data_dir } => {
+            bzod::cli::list_users::run(data_dir, config).await?;
+        }
+        Commands::BackupUser {
+            username,
+            out,
+            data_dir,
+        } => {
+            bzod::cli::backup_user::run(username, out, data_dir, config).await?;
+        }
+        Commands::RestoreUser { file, data_dir } => {
+            bzod::cli::restore_user::run(file, data_dir, config).await?;
+        }
+        Commands::AdminMigrate {
+            target_admin_id,
+            data_dir,
+            dry_run,
+            force,
+        } => {
+            bzod::cli::admin_migrate::run(target_admin_id, data_dir, dry_run, force, config)
+                .await?;
+        }
+        Commands::Repair { command } => {
+            bzod::cli::repair::run(command, config).await?;
+        }
    }

    Ok(())
@@ -9,5 +9,8 @@ pub use api_key::ApiKey;
 pub use audit::AuditLog;
 pub use page::LandingPage;
 pub use url::{AuditEvent, LinkPreview, QrCode, Url};
-pub use user::{Session, User};
+pub use user::{
+    AccountType, ApiActor, ModerationSeverity, Session, SlugStatus, TenantUser, User, UserApiToken,
+    UserQuotas, UserSession, UsernameHistory,
+};
 pub use visit::{SummaryEntry, VisitRecord};
@@ -15,3 +15,193 @@ pub struct Session {
    pub expires_at: String,
    pub created_at: String,
 }
+
+#[derive(Serialize, Deserialize, Clone, Debug)]
+pub struct TenantUser {
+    pub id: i64,
+    pub username: String,
+    pub password_hash: String,
+    pub status: String, // 'active', 'disabled', 'suspended', 'pending', 'deleted'
+    pub created_at: String,
+    pub last_login: Option<String>,
+    pub account_type: String, // 'system', 'admin', 'standard', 'organization', 'service'
+    pub organization_id: Option<i64>,
+    pub metadata: Option<String>,
+}
+
+#[derive(Serialize, Deserialize, Clone, Debug)]
+pub struct UserQuotas {
+    pub user_id: i64,
+    pub max_urls: i64,
+    pub max_landings: i64,
+    pub max_api_tokens: i64,
+    pub max_storage_mb: i64,
+    pub current_urls: i64,
+    pub current_landings: i64,
+    pub current_api_tokens: i64,
+    pub current_storage_mb: i64,
+}
+
+impl UserQuotas {
+    pub fn urls_pct(&self) -> f64 {
+        if self.max_urls <= 0 {
+            0.0
+        } else {
+            (self.current_urls as f64 / self.max_urls as f64 * 100.0).clamp(0.0, 100.0)
+        }
+    }
+    pub fn landings_pct(&self) -> f64 {
+        if self.max_landings <= 0 {
+            0.0
+        } else {
+            (self.current_landings as f64 / self.max_landings as f64 * 100.0).clamp(0.0, 100.0)
+        }
+    }
+    pub fn api_tokens_pct(&self) -> f64 {
+        if self.max_api_tokens <= 0 {
+            0.0
+        } else {
+            (self.current_api_tokens as f64 / self.max_api_tokens as f64 * 100.0).clamp(0.0, 100.0)
+        }
+    }
+    pub fn storage_pct(&self) -> f64 {
+        if self.max_storage_mb <= 0 {
+            0.0
+        } else {
+            (self.current_storage_mb as f64 / self.max_storage_mb as f64 * 100.0).clamp(0.0, 100.0)
+        }
+    }
+}
+
+#[derive(Serialize, Deserialize, Clone, Debug)]
+pub struct UserApiToken {
+    pub id: i64,
+    pub user_id: i64,
+    pub token_hash: String,
+    pub created_at: String,
+}
+
+#[derive(Serialize, Deserialize, Clone, Debug)]
+pub struct UserSession {
+    pub id: String,
+    pub user_id: i64,
+    pub expires_at: String,
+    pub created_at: String,
+}
+
+#[derive(Serialize, Deserialize, Clone, Debug)]
+pub struct UsernameHistory {
+    pub id: i64,
+    pub user_id: i64,
+    pub old_username: String,
+    pub new_username: String,
+    pub changed_at: String,
+}
+
+#[derive(Serialize, Deserialize, Clone, Copy, Debug, PartialEq, Eq)]
+pub enum SlugStatus {
+    Active,
+    Flagged,
+    Disabled,
+    SoftDeleted,
+}
+
+impl SlugStatus {
+    pub fn as_str(&self) -> &'static str {
+        match self {
+            Self::Active => "active",
+            Self::Flagged => "flagged",
+            Self::Disabled => "disabled",
+            Self::SoftDeleted => "soft_deleted",
+        }
+    }
+
+    #[allow(clippy::should_implement_trait)]
+    pub fn from_str(s: &str) -> Option<Self> {
+        match s {
+            "active" => Some(Self::Active),
+            "flagged" => Some(Self::Flagged),
+            "disabled" => Some(Self::Disabled),
+            "soft_deleted" => Some(Self::SoftDeleted),
+            _ => None,
+        }
+    }
+}
+
+#[derive(Serialize, Deserialize, Clone, Copy, Debug, PartialEq, Eq)]
+pub enum AccountType {
+    System,
+    Admin,
+    Standard,
+    Organization,
+    Service,
+}
+
+impl AccountType {
+    pub fn as_str(&self) -> &'static str {
+        match self {
+            Self::System => "system",
+            Self::Admin => "admin",
+            Self::Standard => "standard",
+            Self::Organization => "organization",
+            Self::Service => "service",
+        }
+    }
+
+    #[allow(clippy::should_implement_trait)]
+    pub fn from_str(s: &str) -> Option<Self> {
+        match s {
+            "system" => Some(Self::System),
+            "admin" => Some(Self::Admin),
+            "standard" => Some(Self::Standard),
+            "organization" => Some(Self::Organization),
+            "service" => Some(Self::Service),
+            _ => None,
+        }
+    }
+}
+
+#[derive(Serialize, Deserialize, Clone, Copy, Debug, PartialEq, Eq)]
+pub enum ModerationSeverity {
+    Low,
+    Medium,
+    High,
+    Critical,
+}
+
+impl ModerationSeverity {
+    pub fn as_str(&self) -> &'static str {
+        match self {
+            Self::Low => "low",
+            Self::Medium => "medium",
+            Self::High => "high",
+            Self::Critical => "critical",
+        }
+    }
+
+    #[allow(clippy::should_implement_trait)]
+    pub fn from_str(s: &str) -> Option<Self> {
+        match s {
+            "low" => Some(Self::Low),
+            "medium" => Some(Self::Medium),
+            "high" => Some(Self::High),
+            "critical" => Some(Self::Critical),
+            _ => None,
+        }
+    }
+}
+
+#[derive(Clone, Debug)]
+pub enum ApiActor {
+    Admin(User),
+    User(TenantUser),
+}
+
+impl ApiActor {
+    pub fn username(&self) -> &str {
+        match self {
+            Self::Admin(u) => &u.username,
+            Self::User(u) => &u.username,
+        }
+    }
+}
@@ -12,6 +12,7 @@ pub struct VisitRecord {
    pub accept_language: String,
    pub country: String,
    pub status_code: u16,
+    pub owner_user_id: Option<i64>,
 }

 #[derive(Serialize, Deserialize, Clone, Debug)]
@@ -0,0 +1,121 @@
+//! Post-restore filesystem layout normalization for multi-tenant BZOD data dirs.
+//!
+//! Extracted from admin restore handlers so path moves are testable without HTTP.
+
+use std::path::{Path, PathBuf};
+use tracing::warn;
+
+/// Move flat legacy DB files into multi-tenant paths after tarball extract.
+///
+/// Layout:
+/// - `admin.db` / `system.db` / `users.db` (+ wal/shm) → `{data_dir}/admin/`
+/// - `content.db` / `analytics.db` (+ wal/shm) → `{data_dir}/users/1/`
+pub fn normalize_restored_layout(data_dir: &Path) -> std::io::Result<()> {
+    let admin_dir = data_dir.join("admin");
+    let users_1_dir = data_dir.join("users").join("1");
+    std::fs::create_dir_all(&admin_dir)?;
+    std::fs::create_dir_all(&users_1_dir)?;
+
+    let admin_files = [
+        "admin.db",
+        "admin.db-wal",
+        "admin.db-shm",
+        "system.db",
+        "system.db-wal",
+        "system.db-shm",
+        "users.db",
+        "users.db-wal",
+        "users.db-shm",
+    ];
+    for f in admin_files {
+        let src = data_dir.join(f);
+        if src.exists() {
+            let dst = admin_dir.join(f);
+            if let Err(e) = std::fs::rename(&src, &dst) {
+                warn!(
+                    file = f,
+                    error = %e,
+                    "failed to move restored admin file into admin/"
+                );
+                return Err(e);
+            }
+        }
+    }
+
+    let content_files = [
+        "content.db",
+        "content.db-wal",
+        "content.db-shm",
+        "analytics.db",
+        "analytics.db-wal",
+        "analytics.db-shm",
+    ];
+    for f in content_files {
+        let src = data_dir.join(f);
+        if src.exists() {
+            let dst = users_1_dir.join(f);
+            if let Err(e) = std::fs::rename(&src, &dst) {
+                warn!(
+                    file = f,
+                    error = %e,
+                    "failed to move restored content file into users/1/"
+                );
+                return Err(e);
+            }
+        }
+    }
+
+    Ok(())
+}
+
+/// Paths used when reopening connections after restore.
+#[derive(Debug, Clone)]
+pub struct RestoredDbPaths {
+    pub admin: PathBuf,
+    pub system: PathBuf,
+    pub users: PathBuf,
+    pub content: PathBuf,
+    pub analytics: PathBuf,
+}
+
+impl RestoredDbPaths {
+    pub fn from_data_dir(data_dir: &Path) -> Self {
+        Self {
+            admin: data_dir.join("admin/admin.db"),
+            system: data_dir.join("admin/system.db"),
+            users: data_dir.join("admin/users.db"),
+            content: data_dir.join("users/1/content.db"),
+            analytics: data_dir.join("users/1/analytics.db"),
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use std::fs;
+
+    #[test]
+    fn moves_flat_files_into_tenant_layout() {
+        let dir = std::env::temp_dir().join(format!("bzod_layout_{}", uuid::Uuid::new_v4()));
+        let _ = fs::remove_dir_all(&dir);
+        fs::create_dir_all(&dir).unwrap();
+        fs::write(dir.join("admin.db"), b"a").unwrap();
+        fs::write(dir.join("system.db"), b"s").unwrap();
+        fs::write(dir.join("users.db"), b"u").unwrap();
+        fs::write(dir.join("content.db"), b"c").unwrap();
+        fs::write(dir.join("analytics.db"), b"an").unwrap();
+
+        normalize_restored_layout(&dir).unwrap();
+
+        assert!(dir.join("admin/admin.db").exists());
+        assert!(dir.join("admin/system.db").exists());
+        assert!(dir.join("admin/users.db").exists());
+        assert!(dir.join("users/1/content.db").exists());
+        assert!(dir.join("users/1/analytics.db").exists());
+        assert!(!dir.join("admin.db").exists());
+        assert!(!dir.join("content.db").exists());
+
+        let _ = fs::remove_dir_all(&dir);
+    }
+}
@@ -0,0 +1,204 @@
+//! Bulk URL creation business logic (transaction + slug reservation).
+//!
+//! Handlers own auth/HTTP; this module owns validation, reservation, and inserts.
+
+use crate::auth::generate_token;
+use crate::auth::password::hash_password;
+use crate::models::Url;
+use crate::utils::validation::validate_redirect_destination;
+use rusqlite::{Connection, Transaction};
+use std::sync::Mutex;
+
+/// One item in a bulk URL create request (mirrors the HTTP payload shape).
+#[derive(Debug, Clone)]
+pub struct BulkUrlCreateItem {
+    pub destination: String,
+    pub code: Option<String>,
+    pub title: Option<String>,
+    pub description: Option<String>,
+    pub tags: Option<Vec<String>>,
+    pub expires_at: Option<String>,
+    pub password: Option<String>,
+    pub max_access_count: Option<i64>,
+}
+
+#[derive(Debug)]
+pub enum BulkUrlError {
+    BadRequest(String),
+    Conflict(String),
+    Forbidden(String),
+    Internal(String),
+}
+
+impl BulkUrlError {
+    pub fn message(&self) -> &str {
+        match self {
+            Self::BadRequest(m) | Self::Conflict(m) | Self::Forbidden(m) | Self::Internal(m) => m,
+        }
+    }
+}
+
+fn release_reserved(system: &Connection, slugs: &[String], owner_user_id: i64) {
+    for slug in slugs {
+        let _ = crate::db::users::release_global_slug(system, slug, owner_user_id);
+    }
+}
+
+/// Check that the tenant can accept `additional` new URLs.
+pub fn ensure_url_quota(
+    users_db: &Mutex<Connection>,
+    user_id: i64,
+    additional: i64,
+) -> Result<(), BulkUrlError> {
+    let users_conn = crate::utils::lock_db(users_db, "users_db")
+        .map_err(|e| BulkUrlError::Internal(e.to_string()))?;
+    match crate::db::users::get_user_quotas(&users_conn, user_id) {
+        Ok(Some(quotas)) => {
+            if quotas.current_urls + additional > quotas.max_urls {
+                Err(BulkUrlError::Forbidden("Quota limit exceeded".into()))
+            } else {
+                Ok(())
+            }
+        }
+        Ok(None) => Err(BulkUrlError::Forbidden("User quota not found".into())),
+        Err(e) => Err(BulkUrlError::Internal(format!("quota lookup failed: {e}"))),
+    }
+}
+
+/// Create many URLs inside a single content transaction with global slug reservation.
+pub fn create_urls_bulk(
+    content_db: &Mutex<Connection>,
+    system_db: &Mutex<Connection>,
+    users_db: &Mutex<Connection>,
+    owner_user_id: i64,
+    items: Vec<BulkUrlCreateItem>,
+) -> Result<Vec<Url>, BulkUrlError> {
+    let mut conn = crate::utils::lock_db(content_db, "content_db")
+        .map_err(|e| BulkUrlError::Internal(e.to_string()))?;
+    let tx = conn.transaction().map_err(|e| {
+        BulkUrlError::Internal(format!("Failed to start database transaction: {e}"))
+    })?;
+
+    let mut created_urls = Vec::new();
+    let mut reserved_slugs: Vec<String> = Vec::new();
+
+    for item in items {
+        match create_one_in_tx(&tx, system_db, owner_user_id, item, &mut reserved_slugs) {
+            Ok(url) => created_urls.push(url),
+            Err(e) => {
+                let _ = tx.rollback();
+                if let Ok(system_conn) = crate::utils::lock_db(system_db, "system_db") {
+                    release_reserved(&system_conn, &reserved_slugs, owner_user_id);
+                }
+                return Err(e);
+            }
+        }
+    }
+
+    if let Err(e) = tx.commit() {
+        if let Ok(system_conn) = crate::utils::lock_db(system_db, "system_db") {
+            release_reserved(&system_conn, &reserved_slugs, owner_user_id);
+        }
+        return Err(BulkUrlError::Internal(format!(
+            "Failed to commit transaction: {e}"
+        )));
+    }
+
+    // Activate slugs
+    {
+        let system_conn = crate::utils::lock_db(system_db, "system_db")
+            .map_err(|e| BulkUrlError::Internal(e.to_string()))?;
+        for url in &created_urls {
+            let _ = system_conn.execute(
+                "UPDATE global_slugs SET target_id = ?1, status = 'active', updated_at = ?2 WHERE slug = ?3;",
+                rusqlite::params![url.id, chrono::Utc::now().to_rfc3339(), url.code],
+            );
+        }
+    }
+
+    // Increment quota counters
+    {
+        let users_conn = crate::utils::lock_db(users_db, "users_db")
+            .map_err(|e| BulkUrlError::Internal(e.to_string()))?;
+        for _ in 0..created_urls.len() {
+            let _ = crate::db::users::increment_quota_counter(&users_conn, owner_user_id, "urls");
+        }
+    }
+
+    Ok(created_urls)
+}
+
+fn create_one_in_tx(
+    tx: &Transaction<'_>,
+    system_db: &Mutex<Connection>,
+    owner_user_id: i64,
+    item: BulkUrlCreateItem,
+    reserved_slugs: &mut Vec<String>,
+) -> Result<Url, BulkUrlError> {
+    let mut code = item.code.unwrap_or_default().trim().to_lowercase();
+    if code.is_empty() {
+        code = generate_token(3);
+    } else if code.len() != 6 || !code.chars().all(|c| c.is_ascii_hexdigit()) {
+        return Err(BulkUrlError::BadRequest(format!(
+            "Short code '{code}' must be 6 hex characters"
+        )));
+    }
+
+    {
+        let system_conn = crate::utils::lock_db(system_db, "system_db")
+            .map_err(|e| BulkUrlError::Internal(e.to_string()))?;
+        let available = crate::db::users::is_slug_available(&system_conn, &code).unwrap_or(false)
+            && !reserved_slugs.contains(&code);
+        if !available {
+            return Err(BulkUrlError::Conflict(format!(
+                "Short code '{code}' already exists"
+            )));
+        }
+        if let Err(e) = crate::db::users::register_global_slug(
+            &system_conn,
+            &code,
+            owner_user_id,
+            "url",
+            "",
+            "reserving",
+        ) {
+            return Err(BulkUrlError::Internal(format!(
+                "Failed to reserve slug '{code}': {e}"
+            )));
+        }
+        reserved_slugs.push(code.clone());
+    }
+
+    let password_hash = if let Some(ref pwd) = item.password {
+        match hash_password(pwd) {
+            Ok(h) => Some(h),
+            Err(e) => {
+                return Err(BulkUrlError::Internal(format!(
+                    "Password hashing error: {e}"
+                )));
+            }
+        }
+    } else {
+        None
+    };
+
+    if !validate_redirect_destination(&item.destination) {
+        return Err(BulkUrlError::BadRequest(format!(
+            "Invalid destination for item '{code}': must be a valid http(s) URL without control characters"
+        )));
+    }
+
+    let tags = item.tags.unwrap_or_default();
+    crate::db::content::create_url_extended(
+        tx,
+        &code,
+        &item.destination,
+        item.title.as_deref(),
+        item.description.as_deref(),
+        &tags,
+        item.expires_at.as_deref(),
+        password_hash.as_deref(),
+        item.max_access_count,
+    )
+    .map_err(|e| BulkUrlError::Internal(format!("Database insert error: {e}")))
+}
@@ -0,0 +1,261 @@
+//! Read-only audit of stored redirect destinations.
+//!
+//! Scans tenant content databases and classifies each `urls.destination` using
+//! the same rules as write-path validation. Never rewrites or deletes data.
+
+use crate::db::Db;
+use crate::utils::validation::{classify_redirect_destination, DestinationClass};
+use rusqlite::Connection;
+use std::path::Path;
+use tracing::{error, info, warn};
+
+/// Summary counters for a destination audit run.
+#[derive(Debug, Default, Clone, PartialEq, Eq)]
+pub struct DestinationAuditReport {
+    pub scanned_users: usize,
+    pub total_urls: usize,
+    pub valid_http: usize,
+    pub valid_https: usize,
+    pub invalid: usize,
+    pub control_characters: usize,
+    pub unsupported_scheme: usize,
+    pub malformed: usize,
+    pub empty: usize,
+    pub too_long: usize,
+    pub non_ascii: usize,
+    /// Safe sample of invalid records: (owner_user_id, code, class_label).
+    /// Destination bodies are never included (may contain control chars / secrets).
+    pub invalid_samples: Vec<InvalidDestinationSample>,
+}
+
+#[derive(Debug, Clone, PartialEq, Eq)]
+pub struct InvalidDestinationSample {
+    pub owner_user_id: i64,
+    pub code: String,
+    pub url_id: String,
+    pub class: &'static str,
+    pub destination_len: usize,
+}
+
+const MAX_SAMPLES: usize = 50;
+
+fn class_label(c: DestinationClass) -> &'static str {
+    match c {
+        DestinationClass::ValidHttp => "valid_http",
+        DestinationClass::ValidHttps => "valid_https",
+        DestinationClass::Empty => "empty",
+        DestinationClass::TooLong => "too_long",
+        DestinationClass::ControlCharacters => "control_characters",
+        DestinationClass::NonAscii => "non_ascii",
+        DestinationClass::UnsupportedScheme => "unsupported_scheme",
+        DestinationClass::Malformed => "malformed",
+    }
+}
+
+/// Classify a single destination and update report counters.
+pub fn record_destination(
+    report: &mut DestinationAuditReport,
+    owner_user_id: i64,
+    code: &str,
+    url_id: &str,
+    destination: &str,
+) {
+    report.total_urls += 1;
+    let class = classify_redirect_destination(destination);
+    match class {
+        DestinationClass::ValidHttp => report.valid_http += 1,
+        DestinationClass::ValidHttps => report.valid_https += 1,
+        DestinationClass::Empty => {
+            report.empty += 1;
+            report.invalid += 1;
+        }
+        DestinationClass::TooLong => {
+            report.too_long += 1;
+            report.invalid += 1;
+        }
+        DestinationClass::ControlCharacters => {
+            report.control_characters += 1;
+            report.invalid += 1;
+        }
+        DestinationClass::NonAscii => {
+            report.non_ascii += 1;
+            report.invalid += 1;
+        }
+        DestinationClass::UnsupportedScheme => {
+            report.unsupported_scheme += 1;
+            report.invalid += 1;
+        }
+        DestinationClass::Malformed => {
+            report.malformed += 1;
+            report.invalid += 1;
+        }
+    }
+
+    if !class.is_valid() && report.invalid_samples.len() < MAX_SAMPLES {
+        report.invalid_samples.push(InvalidDestinationSample {
+            owner_user_id,
+            code: code.to_string(),
+            url_id: url_id.to_string(),
+            class: class_label(class),
+            destination_len: destination.len(),
+        });
+    }
+}
+
+/// Scan one content database connection for URL destinations.
+pub fn audit_content_conn(
+    conn: &Connection,
+    owner_user_id: i64,
+    report: &mut DestinationAuditReport,
+) -> rusqlite::Result<()> {
+    let mut stmt = conn.prepare("SELECT id, code, destination FROM urls;")?;
+    let rows = stmt.query_map([], |row| {
+        Ok((
+            row.get::<_, String>(0)?,
+            row.get::<_, String>(1)?,
+            row.get::<_, String>(2)?,
+        ))
+    })?;
+
+    for row in rows {
+        let (id, code, destination) = row?;
+        record_destination(report, owner_user_id, &code, &id, &destination);
+    }
+    Ok(())
+}
+
+fn open_user_content(data_dir: &Path, user_id: i64) -> Result<Connection, rusqlite::Error> {
+    let path = data_dir
+        .join("users")
+        .join(user_id.to_string())
+        .join("content.db");
+    if !path.exists() {
+        return Err(rusqlite::Error::InvalidPath(path));
+    }
+    let conn = Connection::open(path)?;
+    crate::db::sqlite::enable_wal(&conn, "content")?;
+    Ok(conn)
+}
+
+/// Audit all tenant content databases found under the configured data directory.
+///
+/// Read-only: does not modify any records.
+pub fn audit_all_destinations(db: &Db) -> Result<DestinationAuditReport, String> {
+    let mut report = DestinationAuditReport::default();
+
+    let user_ids: Vec<i64> = {
+        let users = db
+            .users
+            .lock()
+            .map_err(|e| format!("users_db lock poisoned: {}", e))?;
+        let mut stmt = users
+            .prepare("SELECT id FROM users;")
+            .map_err(|e| e.to_string())?;
+        let rows = stmt
+            .query_map([], |row| row.get(0))
+            .map_err(|e| e.to_string())?;
+        rows.filter_map(|r| r.ok()).collect()
+    };
+
+    for user_id in user_ids {
+        match open_user_content(&db.data_dir, user_id) {
+            Ok(conn) => {
+                report.scanned_users += 1;
+                if let Err(e) = audit_content_conn(&conn, user_id, &mut report) {
+                    error!(
+                        owner_user_id = user_id,
+                        error = %e,
+                        "destination audit failed for user content.db"
+                    );
+                    return Err(format!("audit user {} content.db: {}", user_id, e));
+                }
+            }
+            Err(rusqlite::Error::InvalidPath(_)) => {
+                // User has no content DB yet — skip.
+            }
+            Err(e) => {
+                warn!(
+                    owner_user_id = user_id,
+                    error = %e,
+                    "could not open user content.db for destination audit"
+                );
+            }
+        }
+    }
+
+    info!(
+        total_urls = report.total_urls,
+        valid = report.valid_http + report.valid_https,
+        invalid = report.invalid,
+        "destination audit complete"
+    );
+    Ok(report)
+}
+
+/// Format a human-readable report for CLI output.
+pub fn format_report(report: &DestinationAuditReport) -> String {
+    let mut out = String::new();
+    out.push_str("BZOD Redirect Destination Audit (read-only)\n");
+    out.push_str("===========================================\n");
+    out.push_str(&format!("Users scanned:        {}\n", report.scanned_users));
+    out.push_str(&format!("Total URLs:           {}\n", report.total_urls));
+    out.push_str(&format!("Valid HTTP:           {}\n", report.valid_http));
+    out.push_str(&format!("Valid HTTPS:          {}\n", report.valid_https));
+    out.push_str(&format!("Invalid (total):      {}\n", report.invalid));
+    out.push_str(&format!(
+        "  control characters: {}\n",
+        report.control_characters
+    ));
+    out.push_str(&format!(
+        "  unsupported scheme: {}\n",
+        report.unsupported_scheme
+    ));
+    out.push_str(&format!("  malformed:          {}\n", report.malformed));
+    out.push_str(&format!("  empty:              {}\n", report.empty));
+    out.push_str(&format!("  too long:           {}\n", report.too_long));
+    out.push_str(&format!("  non-ascii:          {}\n", report.non_ascii));
+
+    if !report.invalid_samples.is_empty() {
+        out.push_str("\nInvalid samples (id/code only; destinations not printed):\n");
+        for s in &report.invalid_samples {
+            out.push_str(&format!(
+                "  user={} code={} id={} class={} dest_len={}\n",
+                s.owner_user_id, s.code, s.url_id, s.class, s.destination_len
+            ));
+        }
+    }
+    out
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn records_control_character_destination() {
+        let mut report = DestinationAuditReport::default();
+        record_destination(
+            &mut report,
+            1,
+            "ab12cd",
+            "id-1",
+            "https://evil.example/\r\nX:1",
+        );
+        assert_eq!(report.total_urls, 1);
+        assert_eq!(report.invalid, 1);
+        assert_eq!(report.control_characters, 1);
+        assert_eq!(report.invalid_samples.len(), 1);
+        assert_eq!(report.invalid_samples[0].class, "control_characters");
+        // Ensure we never store the destination body in the sample.
+        assert!(!format!("{:?}", report.invalid_samples[0]).contains("evil"));
+    }
+
+    #[test]
+    fn records_valid_https() {
+        let mut report = DestinationAuditReport::default();
+        record_destination(&mut report, 1, "ab12cd", "id-1", "https://example.com/ok");
+        assert_eq!(report.valid_https, 1);
+        assert_eq!(report.invalid, 0);
+        assert!(report.invalid_samples.is_empty());
+    }
+}
@@ -1,6 +1,12 @@
 pub mod api_keys;
 pub mod audit;
+pub mod backup_layout;
 pub mod bulk;
+pub mod bulk_urls;
+pub mod destination_audit;
 pub mod landing_pages;
 pub mod qr;
+pub mod registry_validator;
 pub mod shortener;
+pub mod slug_transfer;
+pub mod urls;
@@ -0,0 +1,285 @@
+use rusqlite::Connection;
+use std::path::{Path, PathBuf};
+
+#[derive(Debug, Clone, PartialEq)]
+pub enum RegistryIssueType {
+    DuplicateSlug,
+    InvalidTargetType,
+    InvalidStatus,
+    MissingOwner,
+    MissingDatabase,
+    MissingTarget,
+    StaleReservation,
+    TenantAdminHasIsolatedContent,
+}
+
+#[derive(Debug, Clone)]
+pub struct RegistryIssue {
+    pub slug: String,
+    pub target_type: String,
+    pub owner_user_id: i64,
+    pub database_path: PathBuf,
+    pub target_id: String,
+    pub issue_type: RegistryIssueType,
+    pub description: String,
+}
+
+pub struct RegistryValidator;
+
+impl RegistryValidator {
+    /// Scans the global_slugs registry and returns a list of detected issues.
+    pub fn scan(
+        system_conn: &Connection,
+        users_conn: &Connection,
+        data_dir: &Path,
+        slug_filter: Option<&str>,
+    ) -> Result<Vec<RegistryIssue>, Box<dyn std::error::Error>> {
+        use chrono::{DateTime, Utc};
+        let mut issues = Vec::new();
+
+        // 1. Check duplicate slugs (only if not filtering by single slug)
+        if slug_filter.is_none() {
+            let total_count: i64 =
+                system_conn.query_row("SELECT COUNT(*) FROM global_slugs;", [], |r| r.get(0))?;
+            let distinct_count: i64 = system_conn.query_row(
+                "SELECT COUNT(DISTINCT slug) FROM global_slugs;",
+                [],
+                |r| r.get(0),
+            )?;
+            if total_count != distinct_count {
+                issues.push(RegistryIssue {
+                    slug: "*".to_string(),
+                    target_type: "system".to_string(),
+                    owner_user_id: 0,
+                    database_path: data_dir.join("admin/system.db"),
+                    target_id: "".to_string(),
+                    issue_type: RegistryIssueType::DuplicateSlug,
+                    description: format!(
+                        "Duplicate slugs found in global_slugs table (total rows: {}, distinct slugs: {})",
+                        total_count, distinct_count
+                    ),
+                });
+            }
+        }
+
+        // 2. Scan global slugs
+        let (query, params_string) = if let Some(slug) = slug_filter {
+            (
+                "SELECT slug, owner_user_id, target_type, target_id, created_at, status FROM global_slugs WHERE slug = ?1;",
+                vec![slug.to_string()],
+            )
+        } else {
+            (
+                "SELECT slug, owner_user_id, target_type, target_id, created_at, status FROM global_slugs;",
+                vec![],
+            )
+        };
+
+        let mut stmt = system_conn.prepare(query)?;
+        let mut rows = stmt.query(rusqlite::params_from_iter(params_string))?;
+
+        while let Some(row) = rows.next()? {
+            let slug: String = row.get(0)?;
+            let owner_user_id: i64 = row.get(1)?;
+            let target_type: String = row.get(2)?;
+            let target_id: String = row.get(3)?;
+            let created_at_str: String = row.get(4)?;
+            let status: String = row.get(5)?;
+
+            let content_db_path = if owner_user_id == 1 {
+                data_dir.join("users").join("1").join("content.db")
+            } else {
+                data_dir
+                    .join("users")
+                    .join(owner_user_id.to_string())
+                    .join("content.db")
+            };
+
+            // Target type check
+            if target_type != "url" && target_type != "page" {
+                issues.push(RegistryIssue {
+                    slug: slug.clone(),
+                    target_type: target_type.clone(),
+                    owner_user_id,
+                    database_path: content_db_path.clone(),
+                    target_id: target_id.clone(),
+                    issue_type: RegistryIssueType::InvalidTargetType,
+                    description: format!(
+                        "Slug '{}' has invalid target_type '{}'",
+                        slug, target_type
+                    ),
+                });
+            }
+
+            // Status check
+            if status != "active" && status != "disabled" && status != "reserving" {
+                issues.push(RegistryIssue {
+                    slug: slug.clone(),
+                    target_type: target_type.clone(),
+                    owner_user_id,
+                    database_path: content_db_path.clone(),
+                    target_id: target_id.clone(),
+                    issue_type: RegistryIssueType::InvalidStatus,
+                    description: format!("Slug '{}' has invalid status '{}'", slug, status),
+                });
+            }
+
+            // Check owner
+            let owner_exists: bool = users_conn
+                .query_row(
+                    "SELECT EXISTS(SELECT 1 FROM users WHERE id = ?1);",
+                    [owner_user_id],
+                    |r| r.get(0),
+                )
+                .unwrap_or(false);
+
+            if !owner_exists {
+                issues.push(RegistryIssue {
+                    slug: slug.clone(),
+                    target_type: target_type.clone(),
+                    owner_user_id,
+                    database_path: content_db_path.clone(),
+                    target_id: target_id.clone(),
+                    issue_type: RegistryIssueType::MissingOwner,
+                    description: format!(
+                        "Slug '{}' references missing owner user ID {}",
+                        slug, owner_user_id
+                    ),
+                });
+                continue;
+            }
+
+            // Stale warning check
+            if status == "reserving" {
+                if let Ok(created_at) = DateTime::parse_from_rfc3339(&created_at_str) {
+                    let age = Utc::now().signed_duration_since(created_at.with_timezone(&Utc));
+                    if age > chrono::Duration::try_minutes(15).unwrap_or_default() {
+                        issues.push(RegistryIssue {
+                            slug: slug.clone(),
+                            target_type: target_type.clone(),
+                            owner_user_id,
+                            database_path: content_db_path.clone(),
+                            target_id: target_id.clone(),
+                            issue_type: RegistryIssueType::StaleReservation,
+                            description: format!(
+                                "Reserving slug '{}' has been stale for over 15 minutes",
+                                slug
+                            ),
+                        });
+                    }
+                }
+            }
+
+            // Check target record exists for active / disabled (and reserving with target_id)
+            if status == "active"
+                || status == "disabled"
+                || (status == "reserving" && !target_id.is_empty())
+            {
+                if !content_db_path.exists() {
+                    issues.push(RegistryIssue {
+                        slug: slug.clone(),
+                        target_type: target_type.clone(),
+                        owner_user_id,
+                        database_path: content_db_path.clone(),
+                        target_id: target_id.clone(),
+                        issue_type: RegistryIssueType::MissingDatabase,
+                        description: format!(
+                            "Slug '{}' owner content database does not exist at {:?}",
+                            slug, content_db_path
+                        ),
+                    });
+                } else {
+                    match Connection::open(&content_db_path) {
+                        Ok(conn) => {
+                            let exists = if target_type == "url" {
+                                conn.query_row(
+                                    "SELECT EXISTS(SELECT 1 FROM urls WHERE id = ?1);",
+                                    [&target_id],
+                                    |r| r.get(0),
+                                )
+                                .unwrap_or(false)
+                            } else if target_type == "page" {
+                                conn.query_row(
+                                    "SELECT EXISTS(SELECT 1 FROM landing_pages WHERE id = ?1);",
+                                    [&target_id],
+                                    |r| r.get(0),
+                                )
+                                .unwrap_or(false)
+                            } else {
+                                false
+                            };
+
+                            if !exists {
+                                issues.push(RegistryIssue {
+                                    slug: slug.clone(),
+                                    target_type: target_type.clone(),
+                                    owner_user_id,
+                                    database_path: content_db_path.clone(),
+                                    target_id: target_id.clone(),
+                                    issue_type: RegistryIssueType::MissingTarget,
+                                    description: format!("Slug '{}' (type: '{}', id: '{}') references missing target record in owner's content database", slug, target_type, target_id),
+                                });
+                            }
+                        }
+                        Err(e) => {
+                            issues.push(RegistryIssue {
+                                slug: slug.clone(),
+                                target_type: target_type.clone(),
+                                owner_user_id,
+                                database_path: content_db_path.clone(),
+                                target_id: target_id.clone(),
+                                issue_type: RegistryIssueType::MissingDatabase,
+                                description: format!(
+                                    "Slug '{}' owner content database could not be opened: {}",
+                                    slug, e
+                                ),
+                            });
+                        }
+                    }
+                }
+            }
+        }
+
+        // 3. Admin Content Reverse Consistency Check (Legacy DB)
+        // Check if tenant databases contain content for admin users incorrectly (isolated admin content)
+        if slug_filter.is_none() {
+            let mut stmt = users_conn.prepare(
+                "SELECT id, username FROM users WHERE account_type = 'admin' AND id != 1;",
+            )?;
+            let mut admin_rows = stmt.query([])?;
+            while let Some(row) = admin_rows.next()? {
+                let id: i64 = row.get(0)?;
+                let username: String = row.get(1)?;
+                let tenant_db_path = data_dir
+                    .join("users")
+                    .join(id.to_string())
+                    .join("content.db");
+
+                if tenant_db_path.exists() {
+                    if let Ok(tenant_conn) = Connection::open(&tenant_db_path) {
+                        let url_count: i64 = tenant_conn
+                            .query_row("SELECT COUNT(*) FROM urls;", [], |r| r.get(0))
+                            .unwrap_or(0);
+                        let page_count: i64 = tenant_conn
+                            .query_row("SELECT COUNT(*) FROM landing_pages;", [], |r| r.get(0))
+                            .unwrap_or(0);
+
+                        if url_count > 0 || page_count > 0 {
+                            issues.push(RegistryIssue {
+                                slug: "*".to_string(),
+                                target_type: "system".to_string(),
+                                owner_user_id: id,
+                                database_path: tenant_db_path.clone(),
+                                target_id: "".to_string(),
+                                issue_type: RegistryIssueType::TenantAdminHasIsolatedContent,
+                                description: format!("Admin user '{}' (ID {}) has content in isolated tenant DB ({} URLs, {} pages). Admin content should be in legacy DB 1.", username, id, url_count, page_count),
+                            });
+                        }
+                    }
+                }
+            }
+        }
+
+        Ok(issues)
+    }
+}
@@ -10,13 +10,24 @@ pub fn create_url(
    description: Option<&str>,
    tags: &[String],
 ) -> Result<Url, AppError> {
-    let conn = db.content.lock().unwrap();
+    if !crate::utils::validation::validate_redirect_destination(destination) {
+        return Err(AppError::BadRequest(
+            "Destination must be a valid http(s) URL without control characters".into(),
+        ));
+    }
+    let conn = db
+        .content
+        .lock()
+        .map_err(|e| AppError::Internal(format!("content_db mutex poisoned: {}", e)))?;
    let url = crate::db::content::create_url(&conn, code, destination, title, description, tags)?;
    Ok(url)
 }

 pub fn get_url_by_code(db: &Db, code: &str) -> Result<Option<Url>, AppError> {
-    let conn = db.content.lock().unwrap();
+    let conn = db
+        .content
+        .lock()
+        .map_err(|e| AppError::Internal(format!("content_db mutex poisoned: {}", e)))?;
    let url = crate::db::content::get_url_by_code(&conn, code)?;
    Ok(url)
 }
@@ -0,0 +1,243 @@
+//! Cross-tenant slug transfer business logic.
+//!
+//! Copies URL/page content between tenant content DBs, then updates global_slugs
+//! ownership. Handlers own admin auth and HTTP mapping.
+
+use crate::state::{AppState, UserDbs};
+use crate::utils::lock_db;
+use chrono::Utc;
+use rusqlite::OptionalExtension;
+
+#[derive(Debug)]
+pub enum TransferError {
+    NotFound(&'static str),
+    BadRequest(String),
+    Internal(String),
+}
+
+impl TransferError {
+    pub fn message(&self) -> String {
+        match self {
+            Self::NotFound(m) => (*m).to_string(),
+            Self::BadRequest(m) | Self::Internal(m) => m.clone(),
+        }
+    }
+}
+
+#[derive(Debug, Clone)]
+pub struct SlugTransferRequest {
+    pub slug: String,
+    pub new_owner_user_id: i64,
+}
+
+#[derive(Debug)]
+pub struct SlugTransferResult {
+    pub old_owner_user_id: i64,
+    pub new_owner_user_id: i64,
+    pub target_type: String,
+    pub new_target_id: String,
+}
+
+/// Look up slug ownership in `global_slugs`.
+pub fn lookup_slug(state: &AppState, slug: &str) -> Result<(i64, String, String), TransferError> {
+    let system_conn = lock_db(&state.system_db, "system_db")
+        .map_err(|e| TransferError::Internal(e.to_string()))?;
+    let mut stmt = system_conn
+        .prepare("SELECT owner_user_id, target_type, target_id FROM global_slugs WHERE slug = ?1;")
+        .map_err(|e| TransferError::Internal(e.to_string()))?;
+    let row_opt = stmt
+        .query_row([slug], |row| {
+            Ok((
+                row.get::<_, i64>(0)?,
+                row.get::<_, String>(1)?,
+                row.get::<_, String>(2)?,
+            ))
+        })
+        .optional()
+        .map_err(|e| TransferError::Internal(e.to_string()))?;
+
+    match row_opt {
+        Some(r) => Ok(r),
+        None => Err(TransferError::NotFound("Slug not found")),
+    }
+}
+
+/// Copy content row between tenants and return the new target id.
+fn copy_content(
+    state: &AppState,
+    old_dbs: &UserDbs,
+    new_dbs: &UserDbs,
+    slug: &str,
+    target_type: &str,
+    new_owner_user_id: i64,
+) -> Result<String, TransferError> {
+    let old_conn = lock_db(&old_dbs.content, "old_content_db")
+        .map_err(|e| TransferError::Internal(e.to_string()))?;
+    let new_conn = lock_db(&new_dbs.content, "new_content_db")
+        .map_err(|e| TransferError::Internal(e.to_string()))?;
+
+    if target_type == "url" {
+        let url = match crate::db::content::get_url_by_code(&old_conn, slug) {
+            Ok(Some(u)) => u,
+            Ok(None) => {
+                return Err(TransferError::NotFound(
+                    "Content not found in owner database",
+                ))
+            }
+            Err(e) => return Err(TransferError::Internal(e.to_string())),
+        };
+
+        {
+            let new_users_conn = lock_db(&state.users_db, "users_db")
+                .map_err(|e| TransferError::Internal(e.to_string()))?;
+            if let Ok(Some(quota)) =
+                crate::db::users::get_user_quotas(&new_users_conn, new_owner_user_id)
+            {
+                if quota.current_urls >= quota.max_urls {
+                    return Err(TransferError::BadRequest(
+                        "New owner has exceeded URL quota limit".into(),
+                    ));
+                }
+            }
+        }
+
+        let new_url = crate::db::content::create_url_extended(
+            &new_conn,
+            &url.code,
+            &url.destination,
+            url.title.as_deref(),
+            url.description.as_deref(),
+            &url.tags,
+            url.expires_at.as_deref(),
+            url.password_hash.as_deref(),
+            url.max_access_count,
+        )
+        .map_err(|e| TransferError::Internal(format!("Failed to copy URL to new owner: {e}")))?;
+        let _ = crate::db::content::delete_url(&old_conn, &url.id);
+        Ok(new_url.id)
+    } else if target_type == "page" {
+        let page = match crate::db::content::get_landing_page_by_code(&old_conn, slug) {
+            Ok(Some(p)) => p,
+            Ok(None) => {
+                return Err(TransferError::NotFound(
+                    "Content not found in owner database",
+                ))
+            }
+            Err(e) => return Err(TransferError::Internal(e.to_string())),
+        };
+
+        {
+            let new_users_conn = lock_db(&state.users_db, "users_db")
+                .map_err(|e| TransferError::Internal(e.to_string()))?;
+            if let Ok(Some(quota)) =
+                crate::db::users::get_user_quotas(&new_users_conn, new_owner_user_id)
+            {
+                if quota.current_landings >= quota.max_landings {
+                    return Err(TransferError::BadRequest(
+                        "New owner has exceeded landing page quota limit".into(),
+                    ));
+                }
+            }
+        }
+
+        let new_page = crate::db::content::create_landing_page(
+            &new_conn,
+            &page.code,
+            &page.slug,
+            &page.title,
+            &page.html_content,
+            &page.state,
+        )
+        .map_err(|e| TransferError::Internal(format!("Failed to copy Page to new owner: {e}")))?;
+        let _ = crate::db::content::delete_landing_page(&old_conn, &page.id);
+        Ok(new_page.id)
+    } else {
+        Err(TransferError::NotFound(
+            "Content not found in owner database",
+        ))
+    }
+}
+
+/// Perform a full slug transfer (content + registry + quotas + history).
+pub fn transfer_slug(
+    state: &AppState,
+    req: &SlugTransferRequest,
+    admin_username: &str,
+) -> Result<SlugTransferResult, TransferError> {
+    let (old_owner_user_id, target_type, _target_id) = lookup_slug(state, &req.slug)?;
+
+    if old_owner_user_id == req.new_owner_user_id {
+        return Err(TransferError::BadRequest(
+            "New owner must be different from the current owner".into(),
+        ));
+    }
+
+    let old_dbs = state
+        .get_user_dbs(old_owner_user_id)
+        .map_err(|_| TransferError::Internal("Failed to load current owner's database".into()))?;
+    let new_dbs = state
+        .get_user_dbs(req.new_owner_user_id)
+        .map_err(|_| TransferError::Internal("Failed to load new owner's database".into()))?;
+
+    let new_target_id = copy_content(
+        state,
+        &old_dbs,
+        &new_dbs,
+        &req.slug,
+        &target_type,
+        req.new_owner_user_id,
+    )?;
+
+    {
+        let system_conn = lock_db(&state.system_db, "system_db")
+            .map_err(|e| TransferError::Internal(e.to_string()))?;
+        let now = Utc::now().to_rfc3339();
+
+        let _ = system_conn.execute(
+            "UPDATE global_slugs SET owner_user_id = ?1, target_id = ?2, updated_at = ?3 WHERE slug = ?4;",
+            rusqlite::params![req.new_owner_user_id, new_target_id, now, req.slug],
+        );
+
+        let _ = system_conn.execute(
+            "INSERT INTO slug_history (slug, old_owner_user_id, new_owner_user_id, action, timestamp, admin_username)
+             VALUES (?1, ?2, ?3, 'transferred', ?4, ?5);",
+            rusqlite::params![
+                req.slug,
+                old_owner_user_id,
+                req.new_owner_user_id,
+                now,
+                admin_username
+            ],
+        );
+
+        let users_conn = lock_db(&state.users_db, "users_db")
+            .map_err(|e| TransferError::Internal(e.to_string()))?;
+        let field = if target_type == "url" {
+            "urls"
+        } else {
+            "landings"
+        };
+        let _ = crate::db::users::decrement_quota_counter(&users_conn, old_owner_user_id, field);
+        let _ =
+            crate::db::users::increment_quota_counter(&users_conn, req.new_owner_user_id, field);
+
+        let _ = crate::db::audit_events::write_audit_event(
+            &system_conn,
+            admin_username,
+            "SLUG_TRANSFER",
+            "slug",
+            &req.slug,
+            Some(&format!(
+                "From owner {} to owner {}",
+                old_owner_user_id, req.new_owner_user_id
+            )),
+        );
+    }
+
+    Ok(SlugTransferResult {
+        old_owner_user_id,
+        new_owner_user_id: req.new_owner_user_id,
+        target_type,
+        new_target_id,
+    })
+}
@@ -0,0 +1,117 @@
+//! Shared URL write-path helpers used by admin UI, tenant UI, and REST API.
+//!
+//! Handlers remain responsible for auth/CSRF/quotas; this module owns pure
+//! destination preparation that must stay consistent across entry points.
+
+use crate::utils::validation::validate_redirect_destination;
+
+/// Optional UTM parameters applied to a destination URL.
+#[derive(Debug, Default, Clone)]
+pub struct UtmParams<'a> {
+    pub source: Option<&'a str>,
+    pub medium: Option<&'a str>,
+    pub campaign: Option<&'a str>,
+}
+
+/// Normalize and optionally append UTM parameters to a destination.
+///
+/// Returns `Err` when the base destination fails canonical validation.
+/// UTM appending only runs when the base parses as a URL (same as prior handlers).
+pub fn prepare_destination(raw: &str, utm: UtmParams<'_>) -> Result<String, &'static str> {
+    let mut dest = raw.trim().to_string();
+    if !validate_redirect_destination(&dest) {
+        return Err("Destination must be a valid http(s) URL without control characters");
+    }
+
+    if let Ok(mut parsed) = reqwest::Url::parse(&dest) {
+        let mut has_utm = false;
+        {
+            let mut query = parsed.query_pairs_mut();
+            if let Some(src) = utm.source {
+                let src = src.trim();
+                if !src.is_empty() {
+                    query.append_pair("utm_source", src);
+                    has_utm = true;
+                }
+            }
+            if let Some(med) = utm.medium {
+                let med = med.trim();
+                if !med.is_empty() {
+                    query.append_pair("utm_medium", med);
+                    has_utm = true;
+                }
+            }
+            if let Some(camp) = utm.campaign {
+                let camp = camp.trim();
+                if !camp.is_empty() {
+                    query.append_pair("utm_campaign", camp);
+                    has_utm = true;
+                }
+            }
+        }
+        if has_utm {
+            dest = parsed.to_string();
+        }
+    }
+
+    Ok(dest)
+}
+
+/// Parse HTML datetime-local / partial RFC3339 expiry input into RFC3339 if present.
+pub fn parse_expires_at_input(raw: &str) -> Option<String> {
+    let trimmed = raw.trim();
+    if trimmed.is_empty() {
+        return None;
+    }
+    let mut rfc = trimmed.to_string();
+    if rfc.len() == 16 {
+        // HTML datetime-local → assume UTC seconds
+        rfc.push_str(":00Z");
+    }
+    Some(rfc)
+}
+
+/// Parse optional max-access-count form field.
+pub fn parse_max_access_count(raw: &str) -> Option<i64> {
+    let trimmed = raw.trim();
+    if trimmed.is_empty() {
+        return None;
+    }
+    trimmed.parse().ok()
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn prepare_rejects_crlf() {
+        let err = prepare_destination("https://x/\r\nY:1", UtmParams::default()).unwrap_err();
+        assert!(err.contains("valid http"));
+    }
+
+    #[test]
+    fn prepare_appends_utm() {
+        let dest = prepare_destination(
+            "https://example.com/path",
+            UtmParams {
+                source: Some("newsletter"),
+                medium: Some("email"),
+                campaign: Some("spring"),
+            },
+        )
+        .unwrap();
+        assert!(dest.contains("utm_source=newsletter"));
+        assert!(dest.contains("utm_medium=email"));
+        assert!(dest.contains("utm_campaign=spring"));
+    }
+
+    #[test]
+    fn parse_expires_datetime_local() {
+        assert_eq!(
+            parse_expires_at_input("2030-01-01T12:00"),
+            Some("2030-01-01T12:00:00Z".to_string())
+        );
+        assert_eq!(parse_expires_at_input("  "), None);
+    }
+}
@@ -2,15 +2,25 @@ use crate::analytics::queue::AnalyticsQueue;
 use crate::config::Config;
 use crate::db::Db;
 use rusqlite::Connection;
+use std::collections::HashMap;
 use std::sync::{Arc, Mutex};
 use std::time::Instant;

+#[derive(Clone)]
+pub struct UserDbs {
+    pub content: Arc<Mutex<Connection>>,
+    pub analytics: Arc<Mutex<Connection>>,
+    pub profile: Arc<Mutex<Connection>>,
+}
+
 #[derive(Clone)]
 pub struct AppState {
    pub admin_db: Arc<Mutex<Connection>>,
    pub content_db: Arc<Mutex<Connection>>,
    pub analytics_db: Arc<Mutex<Connection>>,
    pub system_db: Arc<Mutex<Connection>>,
+    pub users_db: Arc<Mutex<Connection>>,
+    pub user_dbs: Arc<Mutex<HashMap<i64, UserDbs>>>,
    pub db: Db,
    pub config: Config,
    pub analytics_queue: AnalyticsQueue,
@@ -18,11 +28,71 @@ pub struct AppState {
 }

 impl AppState {
-    pub fn db_compact(&self) -> Result<(), rusqlite::Error> {
-        self.admin_db.lock().unwrap().execute("VACUUM;", [])?;
-        self.content_db.lock().unwrap().execute("VACUUM;", [])?;
-        self.analytics_db.lock().unwrap().execute("VACUUM;", [])?;
-        self.system_db.lock().unwrap().execute("VACUUM;", [])?;
+    pub fn get_user_dbs(&self, user_id: i64) -> Result<UserDbs, crate::error::AppError> {
+        let mut pool = crate::utils::lock_db(&self.user_dbs, "user_dbs")?;
+        if let Some(dbs) = pool.get(&user_id) {
+            return Ok(dbs.clone());
+        }
+
+        // Open connection and run migrations
+        let user_dir = self.config.data_dir.join("users").join(user_id.to_string());
+        std::fs::create_dir_all(&user_dir)?;
+
+        let content_path = user_dir.join("content.db");
+        let analytics_path = user_dir.join("analytics.db");
+        let profile_path = user_dir.join("profile.db");
+
+        let mut content_conn = Connection::open(content_path)?;
+        let mut analytics_conn = Connection::open(analytics_path)?;
+        let profile_conn = Connection::open(profile_path)?;
+
+        crate::db::sqlite::enable_wal(&content_conn, "content")?;
+        crate::db::sqlite::enable_wal(&analytics_conn, "analytics")?;
+        crate::db::sqlite::enable_wal(&profile_conn, "profile")?;
+
+        crate::db::sqlite::enable_foreign_keys(&content_conn, "content")?;
+        crate::db::sqlite::enable_foreign_keys(&analytics_conn, "analytics")?;
+        crate::db::sqlite::enable_foreign_keys(&profile_conn, "profile")?;
+
+        // Run migrations
+        crate::db::migrations::run_migrations(
+            &mut content_conn,
+            "content",
+            crate::db::migrations::CONTENT_MIGRATIONS,
+            Some(&self.system_db),
+        )
+        .map_err(|e| crate::error::AppError::Internal(e.to_string()))?;
+        crate::db::migrations::run_migrations(
+            &mut analytics_conn,
+            "analytics",
+            crate::db::migrations::ANALYTICS_MIGRATIONS,
+            Some(&self.system_db),
+        )
+        .map_err(|e| crate::error::AppError::Internal(e.to_string()))?;
+
+        profile_conn.execute_batch(
+            "CREATE TABLE IF NOT EXISTS settings (
+                key TEXT PRIMARY KEY,
+                value TEXT NOT NULL
+            );",
+        )?;
+
+        let dbs = UserDbs {
+            content: Arc::new(Mutex::new(content_conn)),
+            analytics: Arc::new(Mutex::new(analytics_conn)),
+            profile: Arc::new(Mutex::new(profile_conn)),
+        };
+
+        pool.insert(user_id, dbs.clone());
+        Ok(dbs)
+    }
+
+    pub fn db_compact(&self) -> Result<(), crate::error::AppError> {
+        crate::utils::lock_db(&self.admin_db, "admin_db")?.execute("VACUUM;", [])?;
+        crate::utils::lock_db(&self.content_db, "content_db")?.execute("VACUUM;", [])?;
+        crate::utils::lock_db(&self.analytics_db, "analytics_db")?.execute("VACUUM;", [])?;
+        crate::utils::lock_db(&self.system_db, "system_db")?.execute("VACUUM;", [])?;
+        crate::utils::lock_db(&self.users_db, "users_db")?.execute("VACUUM;", [])?;
        Ok(())
    }
 }
@@ -0,0 +1,108 @@
+use crate::models::{LandingPage, Url};
+use askama::Template;
+use axum::{
+    http::StatusCode,
+    response::{Html, IntoResponse, Response},
+};
+
+#[derive(Clone, Debug)]
+pub struct VisitorLogEntry {
+    pub sr: usize,
+    pub timestamp: String,
+    pub ip_address: String,
+    pub country: String,
+    pub referrer: String,
+    pub browser: String,
+    pub user_agent: String,
+    pub utm_source: String,
+    pub utm_campaign: String,
+}
+
+#[derive(Template)]
+#[template(path = "url_analytics.html")]
+pub struct UrlAnalyticsTemplate {
+    pub admin_username: String,
+    pub url: Url,
+    pub total_clicks: i64,
+    pub unique_visitors: i64,
+    pub qr_scans: i64,
+    pub direct_clicks: i64,
+    pub traffic_chart: String,
+    pub monthly_chart: String,
+    pub countries_chart: String,
+    pub referrers_chart: String,
+    pub browsers_chart: String,
+    // Paginated visitor logs
+    pub visits: Vec<VisitorLogEntry>,
+    pub current_page: usize,
+    pub total_pages: usize,
+    pub visible_pages: Vec<usize>,
+    pub total_records: i64,
+    pub page_start: usize,
+    pub page_end: usize,
+    pub date_from: Option<String>,
+    pub date_to: Option<String>,
+    pub is_admin: bool,
+}
+
+impl UrlAnalyticsTemplate {
+    pub fn is_current(&self, page: &usize) -> bool {
+        *page == self.current_page
+    }
+}
+
+impl IntoResponse for UrlAnalyticsTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "page_analytics.html")]
+pub struct PageAnalyticsTemplate {
+    pub admin_username: String,
+    pub page: LandingPage,
+    pub total_views: i64,
+    pub unique_visitors: i64,
+    pub traffic_chart: String,
+    pub monthly_chart: String,
+    pub countries_chart: String,
+    pub referrers_chart: String,
+    // Paginated visitor logs
+    pub visits: Vec<VisitorLogEntry>,
+    pub current_page: usize,
+    pub total_pages: usize,
+    pub visible_pages: Vec<usize>,
+    pub total_records: i64,
+    pub page_start: usize,
+    pub page_end: usize,
+    pub date_from: Option<String>,
+    pub date_to: Option<String>,
+    pub is_admin: bool,
+}
+
+impl PageAnalyticsTemplate {
+    pub fn is_current(&self, page: &usize) -> bool {
+        *page == self.current_page
+    }
+}
+
+impl IntoResponse for PageAnalyticsTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
@@ -1,26 +1,41 @@
+pub mod analytics;
 pub mod dashboard;
 pub mod pages;
 pub mod settings;
 pub mod stats;
 pub mod urls;
+pub mod user_dashboard;
+pub mod user_pages;
+pub mod user_settings;
+pub mod user_urls;
+pub mod users;

-pub use dashboard::DashboardTemplate;
-pub use pages::PagesTemplate;
-pub use settings::SettingsTemplate;
-pub use stats::{AuditTemplate, StatusTemplate};
-pub use urls::UrlsTemplate;
-
+pub use analytics::{PageAnalyticsTemplate, UrlAnalyticsTemplate, VisitorLogEntry};
 use askama::Template;
 use axum::{
    http::StatusCode,
    response::{Html, IntoResponse, Response},
 };
+pub use dashboard::DashboardTemplate;
+pub use pages::PagesTemplate;
+pub use settings::SettingsTemplate;
+pub use stats::{AuditTemplate, StatusTemplate, UserAuditTemplate, UserStatusTemplate};
+pub use urls::UrlsTemplate;
+pub use user_dashboard::UserDashboardTemplate;
+pub use user_pages::UserPagesTemplate;
+pub use user_settings::UserSettingsTemplate;
+pub use user_urls::UserUrlsTemplate;
+pub use users::UsersTemplate;

 #[derive(Template)]
 #[template(path = "login.html")]
 pub struct LoginTemplate {
    pub error: Option<String>,
    pub csrf_token: String,
+    pub action: String,
+    pub title: String,
+    pub subtitle: String,
+    pub button_text: String,
 }

 impl IntoResponse for LoginTemplate {
@@ -79,3 +94,283 @@ impl IntoResponse for PreviewTemplate {
        }
    }
 }
+
+#[derive(Template)]
+#[template(path = "users_new.html")]
+pub struct UsersNewTemplate {
+    pub admin_username: String,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for UsersNewTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "user_detail.html")]
+pub struct UserDetailTemplate {
+    pub admin_username: String,
+    pub target_user: crate::models::TenantUser,
+    pub stats: crate::web::admin::UserDetailStats,
+    pub sessions: Vec<crate::models::UserSession>,
+    pub tokens: Vec<crate::models::UserApiToken>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for UserDetailTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "user_edit.html")]
+pub struct UserEditTemplate {
+    pub admin_username: String,
+    pub target_user: crate::models::TenantUser,
+    pub quotas: crate::models::UserQuotas,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for UserEditTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "moderation.html")]
+pub struct ModerationTemplate {
+    pub admin_username: String,
+    pub flagged_items: Vec<crate::web::admin::GlobalSlugRow>,
+    pub logs: Vec<crate::web::admin::ModerationLogEntry>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for ModerationTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "slugs.html")]
+pub struct SlugsTemplate {
+    pub admin_username: String,
+    pub slugs: Vec<crate::web::admin::GlobalSlugRow>,
+    pub history: Vec<crate::web::admin::SlugHistoryRow>,
+    pub csrf_token: String,
+    pub search_filter: Option<String>,
+    pub owner_filter: Option<i64>,
+    pub status_filter: Option<String>,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for SlugsTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "sessions.html")]
+pub struct SessionsTemplate {
+    pub admin_username: String,
+    pub sessions: Vec<crate::models::UserSession>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for SessionsTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "quotas.html")]
+pub struct QuotasTemplate {
+    pub admin_username: String,
+    pub quotas: Vec<crate::models::UserQuotas>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for QuotasTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "health.html")]
+pub struct HealthTemplate {
+    pub admin_username: String,
+    pub db_reports: Vec<crate::db::sqlite::DatabaseHealthReport>,
+    pub total_data_size: String,
+    pub system_db_size: String,
+    pub users_db_size: String,
+    pub admin_db_size: String,
+    pub tenants_db_size: String,
+    pub job_history: Vec<crate::web::admin::JobHistoryRow>,
+    pub health_checks: Vec<crate::web::admin::HealthCheckRow>,
+    pub registry_errors: Vec<String>,
+    pub registry_warnings: Vec<String>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for HealthTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "backups.html")]
+pub struct BackupsTemplate {
+    pub admin_username: String,
+    pub files: Vec<crate::web::admin::BackupFileRow>,
+    pub history: Vec<crate::web::admin::BackupHistoryRow>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for BackupsTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "api_tokens.html")]
+pub struct ApiTokensTemplate {
+    pub admin_username: String,
+    pub username: String,
+    pub tokens: Vec<crate::models::UserApiToken>,
+    pub new_token: Option<String>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for ApiTokensTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+#[derive(Template)]
+#[template(path = "user_analytics.html")]
+pub struct UserAnalyticsTemplate {
+    pub admin_username: String,
+    pub username: String,
+    pub total_clicks: i64,
+    pub unique_visitors: i64,
+    pub direct_clicks: i64,
+    pub referred_clicks: i64,
+    pub referrers_chart: String,
+    pub browsers_chart: String,
+    pub visits: Vec<crate::models::VisitRecord>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for UserAnalyticsTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
@@ -12,6 +12,15 @@ pub struct PagesTemplate {
    pub pages: Vec<LandingPage>,
    pub csrf_token: String,
    pub error: Option<String>,
+    pub current_page: usize,
+    pub total_pages: usize,
+    pub visible_pages: Vec<usize>,
+}
+
+impl PagesTemplate {
+    pub fn is_current(&self, page: &usize) -> bool {
+        *page == self.current_page
+    }
 }

 impl IntoResponse for PagesTemplate {
@@ -51,3 +51,50 @@ impl IntoResponse for AuditTemplate {
        }
    }
 }
+
+#[derive(Template)]
+#[template(path = "user_status.html")]
+pub struct UserStatusTemplate {
+    pub admin_username: String,
+    pub app_status: &'static str,
+    pub db_status: String,
+    pub queue_size: usize,
+    pub memory_usage: String,
+    pub uptime: String,
+    pub version: &'static str,
+    pub git_commit: &'static str,
+    pub urls: Vec<crate::models::Url>,
+}
+
+#[derive(Template)]
+#[template(path = "user_audit.html")]
+pub struct UserAuditTemplate {
+    pub admin_username: String,
+    pub logs: Vec<AuditLog>,
+}
+
+impl IntoResponse for UserStatusTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
+
+impl IntoResponse for UserAuditTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
@@ -14,6 +14,15 @@ pub struct UrlsTemplate {
    pub error: Option<String>,
    pub tag_filter: Option<String>,
    pub base_url: String,
+    pub current_page: usize,
+    pub total_pages: usize,
+    pub visible_pages: Vec<usize>,
+}
+
+impl UrlsTemplate {
+    pub fn is_current(&self, page: &usize) -> bool {
+        *page == self.current_page
+    }
 }

 impl IntoResponse for UrlsTemplate {
@@ -0,0 +1,33 @@
+use askama::Template;
+use axum::{
+    http::StatusCode,
+    response::{Html, IntoResponse, Response},
+};
+
+#[derive(Template)]
+#[template(path = "user_dashboard.html")]
+pub struct UserDashboardTemplate {
+    pub admin_username: String,
+    pub total_urls: i64,
+    pub total_pages: i64,
+    pub total_clicks: i64,
+    pub active_links: i64,
+    pub dead_links: i64,
+    pub traffic_chart: String,
+    pub countries_chart: String,
+    pub browsers_chart: String,
+    pub referrers_chart: String,
+}
+
+impl IntoResponse for UserDashboardTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
@@ -0,0 +1,38 @@
+use crate::models::LandingPage;
+use askama::Template;
+use axum::{
+    http::StatusCode,
+    response::{Html, IntoResponse, Response},
+};
+
+#[derive(Template)]
+#[template(path = "user_pages.html")]
+pub struct UserPagesTemplate {
+    pub admin_username: String,
+    pub username: String,
+    pub pages: Vec<LandingPage>,
+    pub csrf_token: String,
+    pub error: Option<String>,
+    pub current_page: usize,
+    pub total_pages: usize,
+    pub visible_pages: Vec<usize>,
+}
+
+impl UserPagesTemplate {
+    pub fn is_current(&self, page: &usize) -> bool {
+        *page == self.current_page
+    }
+}
+
+impl IntoResponse for UserPagesTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
@@ -0,0 +1,28 @@
+use askama::Template;
+use axum::{
+    http::StatusCode,
+    response::{Html, IntoResponse, Response},
+};
+
+#[derive(Template)]
+#[template(path = "user_settings.html")]
+pub struct UserSettingsTemplate {
+    pub admin_username: String,
+    pub username: String,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for UserSettingsTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
@@ -0,0 +1,40 @@
+use crate::models::Url;
+use askama::Template;
+use axum::{
+    http::StatusCode,
+    response::{Html, IntoResponse, Response},
+};
+
+#[derive(Template)]
+#[template(path = "user_urls.html")]
+pub struct UserUrlsTemplate {
+    pub admin_username: String,
+    pub username: String,
+    pub urls: Vec<Url>,
+    pub csrf_token: String,
+    pub error: Option<String>,
+    pub tag_filter: Option<String>,
+    pub base_url: String,
+    pub current_page: usize,
+    pub total_pages: usize,
+    pub visible_pages: Vec<usize>,
+}
+
+impl UserUrlsTemplate {
+    pub fn is_current(&self, page: &usize) -> bool {
+        *page == self.current_page
+    }
+}
+
+impl IntoResponse for UserUrlsTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
@@ -0,0 +1,26 @@
+use crate::models::TenantUser;
+use askama::Template;
+use axum::response::{Html, IntoResponse, Response};
+
+#[derive(Template)]
+#[template(path = "users.html")]
+pub struct UsersTemplate {
+    pub admin_username: String,
+    pub users: Vec<TenantUser>,
+    pub csrf_token: String,
+    pub success: Option<String>,
+    pub error: Option<String>,
+}
+
+impl IntoResponse for UsersTemplate {
+    fn into_response(self) -> Response {
+        match self.render() {
+            Ok(html) => Html(html).into_response(),
+            Err(e) => (
+                axum::http::StatusCode::INTERNAL_SERVER_ERROR,
+                format!("Render error: {}", e),
+            )
+                .into_response(),
+        }
+    }
+}
@@ -0,0 +1,61 @@
+//! Poison-safe helpers for `std::sync::Mutex` around SQLite connections.
+//!
+//! Prefer these on request paths so a poisoned mutex returns a controlled error
+//! instead of panicking the worker thread.
+
+use crate::error::AppError;
+use std::sync::{Mutex, MutexGuard};
+use tracing::error;
+
+/// Acquire a database mutex, mapping poison to [`AppError::Internal`].
+///
+/// Logs the mutex name (not connection contents or secrets).
+pub fn lock_db<'a, T>(
+    mutex: &'a Mutex<T>,
+    name: &'static str,
+) -> Result<MutexGuard<'a, T>, AppError> {
+    mutex.lock().map_err(|e| {
+        error!(mutex = name, error = %e, "database mutex poisoned");
+        AppError::Internal(format!("{name} mutex poisoned"))
+    })
+}
+
+/// Acquire a database mutex, mapping poison to a plain error string.
+///
+/// Useful for handlers that return `(StatusCode, String)` rather than `AppError`.
+pub fn lock_db_str<'a, T>(
+    mutex: &'a Mutex<T>,
+    name: &'static str,
+) -> Result<MutexGuard<'a, T>, String> {
+    mutex.lock().map_err(|e| {
+        error!(mutex = name, error = %e, "database mutex poisoned");
+        format!("{name} mutex poisoned")
+    })
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use std::sync::Mutex;
+
+    #[test]
+    fn lock_db_succeeds_on_healthy_mutex() {
+        let m = Mutex::new(42);
+        let g = lock_db(&m, "test").unwrap();
+        assert_eq!(*g, 42);
+    }
+
+    #[test]
+    fn lock_db_maps_poison() {
+        let m = Mutex::new(1);
+        let _ = std::panic::catch_unwind(|| {
+            let _g = m.lock().unwrap();
+            panic!("poison");
+        });
+        let err = lock_db(&m, "poisoned_db").unwrap_err();
+        match err {
+            AppError::Internal(msg) => assert!(msg.contains("poisoned_db")),
+            other => panic!("unexpected {other:?}"),
+        }
+    }
+}
@@ -1,11 +1,14 @@
+pub mod db_lock;
 pub mod hashing;
 pub mod network;
 pub mod random;
 pub mod system;
 pub mod time;
+pub mod validation;

+pub use db_lock::{lock_db, lock_db_str};
 pub use hashing::sha256_hash;
-pub use network::get_client_ip;
+pub use network::{get_client_ip, resolve_cookie_secure};
 pub use random::generate_token;
 pub use system::{get_db_file_info, get_memory_usage};
 pub use time::format_duration;
@@ -22,3 +22,119 @@ pub fn get_client_ip(headers: &HeaderMap, connect_info: Option<ConnectInfo<Socke
    }
    "127.0.0.1".to_string()
 }
+
+// Helper to safely extract hostname from a Host header, handling IPv6 and ports.
+pub(crate) fn extract_hostname(host_header: &str) -> &str {
+    if host_header.starts_with('[') {
+        if let Some(end_idx) = host_header.find(']') {
+            return &host_header[1..end_idx];
+        }
+    }
+    host_header.split(':').next().unwrap_or(host_header)
+}
+
+/// Determines whether to set the `Secure` flag on a cookie based on deployment context.
+///
+/// Policy:
+/// 1. If X-Forwarded-Proto is explicitly "https", always enforce Secure=true.
+///    (We assume X-Forwarded-Proto is from a trusted proxy. Forged "https" only makes
+///    the cookie safer. We never weaken based on X-Forwarded-Proto=http).
+/// 2. If the exact Host is a local loopback (localhost, 127.0.0.1, ::1) and we are not
+///    explicitly proxied via HTTPS, disable Secure. This prevents browsers from dropping
+///    the cookie during local development over cleartext HTTP.
+/// 3. Otherwise, fall back to the global `cookie_secure` config (which defaults to true
+///    to keep production secure-by-default even if the proxy strips X-Forwarded-Proto).
+pub fn resolve_cookie_secure(config_secure: bool, headers: &HeaderMap) -> bool {
+    // 1. Explicit HTTPS via reverse proxy
+    if let Some(proto) = headers
+        .get("x-forwarded-proto")
+        .and_then(|v| v.to_str().ok())
+    {
+        if proto.eq_ignore_ascii_case("https") {
+            return true;
+        }
+    }
+
+    // 2. Exact loopback development (prevent cookie drop)
+    if let Some(host_hdr) = headers.get("host").and_then(|h| h.to_str().ok()) {
+        let hostname = extract_hostname(host_hdr);
+        if matches!(hostname, "localhost" | "127.0.0.1" | "::1") {
+            return false;
+        }
+    }
+
+    // 3. Global config (normally true)
+    config_secure
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use axum::http::HeaderValue;
+
+    #[test]
+    fn test_extract_hostname() {
+        assert_eq!(extract_hostname("localhost"), "localhost");
+        assert_eq!(extract_hostname("localhost:8080"), "localhost");
+        assert_eq!(extract_hostname("127.0.0.1"), "127.0.0.1");
+        assert_eq!(extract_hostname("127.0.0.1:8080"), "127.0.0.1");
+        assert_eq!(extract_hostname("[::1]"), "::1");
+        assert_eq!(extract_hostname("[::1]:8080"), "::1");
+        assert_eq!(extract_hostname("example.com"), "example.com");
+        assert_eq!(extract_hostname("example.com:443"), "example.com");
+        assert_eq!(
+            extract_hostname("localhost.example.com"),
+            "localhost.example.com"
+        );
+    }
+
+    #[test]
+    fn test_resolve_cookie_secure() {
+        let mut headers = HeaderMap::new();
+
+        // No headers, global config is true -> Secure=true
+        assert!(resolve_cookie_secure(true, &headers));
+        // No headers, global config is false -> Secure=false
+        assert!(!resolve_cookie_secure(false, &headers));
+
+        // Exact loopback matches -> Secure=false
+        let loopback_hosts = [
+            "localhost",
+            "localhost:8080",
+            "127.0.0.1",
+            "127.0.0.1:8080",
+            "[::1]",
+            "[::1]:8080",
+        ];
+        for host in loopback_hosts {
+            headers.insert("host", HeaderValue::from_static(host));
+            assert!(
+                !resolve_cookie_secure(true, &headers),
+                "Failed for host: {}",
+                host
+            );
+        }
+
+        // Non-loopback localhost subdomains -> Secure=true (relying on config)
+        let public_hosts = ["localhost.example.com", "127.0.0.2", "example.com"];
+        for host in public_hosts {
+            headers.insert("host", HeaderValue::from_static(host));
+            assert!(
+                resolve_cookie_secure(true, &headers),
+                "Failed for host: {}",
+                host
+            );
+        }
+
+        // X-Forwarded-Proto = https overrides loopback
+        headers.insert("host", HeaderValue::from_static("localhost"));
+        headers.insert("x-forwarded-proto", HeaderValue::from_static("https"));
+        assert!(resolve_cookie_secure(true, &headers));
+        assert!(resolve_cookie_secure(false, &headers)); // Overrides false config too
+
+        // X-Forwarded-Proto = http DOES NOT override global config
+        headers.insert("host", HeaderValue::from_static("example.com"));
+        headers.insert("x-forwarded-proto", HeaderValue::from_static("http"));
+        assert!(resolve_cookie_secure(true, &headers)); // Still true because of config
+    }
+}
@@ -19,10 +19,11 @@ pub fn get_memory_usage() -> String {
 pub fn get_db_file_info(data_dir: &Path) -> String {
    let mut stats = String::new();
    let files = vec![
-        ("admin.db", "Admin DB"),
-        ("content.db", "Content DB"),
-        ("analytics.db", "Analytics DB"),
-        ("system.db", "System DB"),
+        ("admin/admin.db", "Admin DB"),
+        ("admin/system.db", "System DB"),
+        ("admin/users.db", "Users DB"),
+        ("users/1/content.db", "Legacy Content DB"),
+        ("users/1/analytics.db", "Legacy Analytics DB"),
    ];

    for (f, name) in files {
@@ -0,0 +1,160 @@
+pub fn validate_custom_slug(slug: &str) -> bool {
+    if !slug.starts_with('!') {
+        return false;
+    }
+    let rest = &slug[1..];
+    if rest.is_empty() || rest.len() > 24 {
+        return false;
+    }
+    rest.chars()
+        .all(|c| c.is_ascii_lowercase() || c.is_ascii_digit() || c == '-' || c == '_')
+}
+
+pub fn validate_redirect_code(code: &str) -> bool {
+    (code.len() == 6 && code.chars().all(|c| c.is_ascii_hexdigit())) || validate_custom_slug(code)
+}
+
+pub fn validate_page_code(code: &str) -> bool {
+    (code.len() == 4 && code.chars().all(|c| c.is_ascii_hexdigit())) || validate_custom_slug(code)
+}
+
+/// Classification of a stored or proposed redirect destination.
+///
+/// Used by write-path validation and read-only legacy data audits. Order of checks
+/// matches `validate_redirect_destination` so both share the same rules.
+#[derive(Debug, Clone, Copy, PartialEq, Eq)]
+pub enum DestinationClass {
+    ValidHttp,
+    ValidHttps,
+    Empty,
+    TooLong,
+    ControlCharacters,
+    NonAscii,
+    UnsupportedScheme,
+    Malformed,
+}
+
+impl DestinationClass {
+    pub fn is_valid(self) -> bool {
+        matches!(self, Self::ValidHttp | Self::ValidHttps)
+    }
+}
+
+fn scheme_prefix(dest: &str) -> Option<&str> {
+    let end = dest.find(':')?;
+    let scheme = &dest[..end];
+    if scheme.is_empty() {
+        return None;
+    }
+    if scheme
+        .chars()
+        .all(|c| c.is_ascii_alphanumeric() || c == '+' || c == '.' || c == '-')
+    {
+        Some(scheme)
+    } else {
+        None
+    }
+}
+
+/// Classify a destination using the same rules as write-path validation.
+pub fn classify_redirect_destination(destination: &str) -> DestinationClass {
+    let dest = destination.trim();
+    if dest.is_empty() {
+        return DestinationClass::Empty;
+    }
+    if dest.len() > 2048 {
+        return DestinationClass::TooLong;
+    }
+    // HTTP header / response-splitting: no CR, LF, NUL, or other ASCII controls.
+    if dest.bytes().any(|b| b < 0x20 || b == 0x7f) {
+        return DestinationClass::ControlCharacters;
+    }
+    // HeaderValue also rejects non-visible ASCII in some cases; require pure ASCII.
+    if !dest.is_ascii() {
+        return DestinationClass::NonAscii;
+    }
+    // Reject non-http(s) schemes even when Url::parse fails (e.g. javascript:).
+    if let Some(scheme) = scheme_prefix(dest) {
+        let scheme_l = scheme.to_ascii_lowercase();
+        if scheme_l != "http" && scheme_l != "https" {
+            return DestinationClass::UnsupportedScheme;
+        }
+    }
+    match reqwest::Url::parse(dest) {
+        Ok(url) => {
+            if url.host_str().is_none() {
+                return DestinationClass::Malformed;
+            }
+            match url.scheme() {
+                "http" => DestinationClass::ValidHttp,
+                "https" => DestinationClass::ValidHttps,
+                _ => DestinationClass::UnsupportedScheme,
+            }
+        }
+        Err(_) => DestinationClass::Malformed,
+    }
+}
+
+/// Returns true if `destination` is safe to store and emit as an HTTP Location value.
+///
+/// Rejects CR/LF and other ASCII control characters (response-splitting), empty values,
+/// and non-http(s) schemes. The redirect handler remains defensive even if invalid
+/// values already exist in older data.
+pub fn validate_redirect_destination(destination: &str) -> bool {
+    classify_redirect_destination(destination).is_valid()
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn rejects_crlf_destination() {
+        assert!(!validate_redirect_destination(
+            "https://example.com/\r\nX-Injected: 1"
+        ));
+        assert!(!validate_redirect_destination(
+            "https://example.com/\nX-Injected: 1"
+        ));
+    }
+
+    #[test]
+    fn rejects_non_http_schemes() {
+        assert!(!validate_redirect_destination("javascript:alert(1)"));
+        assert!(!validate_redirect_destination("data:text/html,hi"));
+        assert!(!validate_redirect_destination("/relative/path"));
+    }
+
+    #[test]
+    fn accepts_normal_https() {
+        assert!(validate_redirect_destination(
+            "https://example.com/path?q=1#frag"
+        ));
+        assert!(validate_redirect_destination("http://localhost:8080/x"));
+    }
+
+    #[test]
+    fn classifies_destination_categories() {
+        assert_eq!(
+            classify_redirect_destination("https://ok.example/"),
+            DestinationClass::ValidHttps
+        );
+        assert_eq!(
+            classify_redirect_destination("http://ok.example/"),
+            DestinationClass::ValidHttp
+        );
+        assert_eq!(
+            classify_redirect_destination("javascript:alert(1)"),
+            DestinationClass::UnsupportedScheme
+        );
+        assert_eq!(
+            classify_redirect_destination("https://x/\r\nX:1"),
+            DestinationClass::ControlCharacters
+        );
+        assert_eq!(
+            classify_redirect_destination("not a url"),
+            DestinationClass::Malformed
+        );
+        assert_eq!(classify_redirect_destination(""), DestinationClass::Empty);
+    }
+}
@@ -0,0 +1,236 @@
+use super::*;
+
+#[derive(Deserialize)]
+pub struct CreateApiKeyForm {
+    pub key_name: String,
+    pub csrf_token: String,
+}
+
+// POST /admin/settings/api-keys/create
+pub async fn create_api_key_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    headers: HeaderMap,
+    connect_info: Option<ConnectInfo<SocketAddr>>,
+    Form(form): Form<CreateApiKeyForm>,
+) -> Response {
+    let (user, session_id) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    if !verify_csrf(&session_id, &form.csrf_token) {
+        return Redirect::to("/admin/settings?error=Invalid CSRF token").into_response();
+    }
+
+    let ip = get_client_ip(&headers, connect_info);
+    let key_secret = format!("bzo_{}", generate_token(16));
+
+    use sha2::{Digest, Sha256};
+    let mut hasher = Sha256::new();
+    hasher.update(key_secret.as_bytes());
+    let hashed_key = hex::encode(hasher.finalize());
+
+    let conn = state.admin_db.lock().unwrap();
+    match create_api_key(&conn, &user.id, &form.key_name, &hashed_key) {
+        Ok(api_key) => {
+            let _ = write_audit_log(
+                &conn,
+                &state,
+                &user.username,
+                "API_KEY_CREATED",
+                Some("api_key"),
+                Some(&api_key.id),
+                Some(&ip),
+                headers.get("user-agent").and_then(|h| h.to_str().ok()),
+            );
+            Redirect::to(&format!(
+                "/admin/settings?success=Token generated successfully. **IMPORTANT: Copy your token now, it will never be shown again!** Token value: {}",
+                key_secret
+            )).into_response()
+        }
+        Err(e) => {
+            Redirect::to(&format!("/admin/settings?error=Database error: {}", e)).into_response()
+        }
+    }
+}
+
+// POST /admin/settings/api-keys/revoke/:id
+pub async fn revoke_api_key_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    headers: HeaderMap,
+    connect_info: Option<ConnectInfo<SocketAddr>>,
+    Path(id): Path<String>,
+    Form(form): Form<std::collections::HashMap<String, String>>,
+) -> Response {
+    let (user, session_id) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let csrf_token = form.get("csrf_token").cloned().unwrap_or_default();
+    if !verify_csrf(&session_id, &csrf_token) {
+        return Redirect::to("/admin/settings?error=Invalid CSRF token").into_response();
+    }
+
+    let ip = get_client_ip(&headers, connect_info);
+
+    let conn = state.admin_db.lock().unwrap();
+    match delete_api_key(&conn, &id) {
+        Ok(_) => {
+            let _ = write_audit_log(
+                &conn,
+                &state,
+                &user.username,
+                "API_KEY_REVOKED",
+                Some("api_key"),
+                Some(&id),
+                Some(&ip),
+                headers.get("user-agent").and_then(|h| h.to_str().ok()),
+            );
+            Redirect::to("/admin/settings?success=API Token revoked").into_response()
+        }
+        Err(e) => Redirect::to(&format!(
+            "/admin/settings?error=Failed to revoke key: {}",
+            e
+        ))
+        .into_response(),
+    }
+}
+
+// GET /api-tokens
+pub async fn api_tokens_get(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    Query(params): Query<HashMap<String, String>>,
+) -> Response {
+    let (user, session_id) = match require_user_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let tokens = {
+        let conn = state.users_db.lock().unwrap();
+        let mut stmt = conn
+            .prepare(
+                "SELECT id, user_id, token_hash, created_at FROM api_tokens WHERE user_id = ?1;",
+            )
+            .unwrap();
+        let rows = stmt
+            .query_map([user.id], |row| {
+                Ok(crate::models::UserApiToken {
+                    id: row.get(0)?,
+                    user_id: row.get(1)?,
+                    token_hash: row.get(2)?,
+                    created_at: row.get(3)?,
+                })
+            })
+            .unwrap();
+        rows.filter_map(|r| r.ok()).collect()
+    };
+
+    let csrf_token = generate_csrf_token(&session_id);
+
+    let template = crate::templates::ApiTokensTemplate {
+        admin_username: user.username.clone(),
+        username: user.username,
+        tokens,
+        new_token: params.get("new_token").cloned(),
+        csrf_token,
+        success: params.get("success").cloned(),
+        error: params.get("error").cloned(),
+    };
+
+    template.into_response()
+}
+
+// POST /api-tokens/create
+pub async fn api_tokens_create_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    Form(form): Form<HashMap<String, String>>,
+) -> Response {
+    let (user, session_id) = match require_user_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let form_csrf = form.get("csrf_token").cloned().unwrap_or_default();
+    if !verify_csrf(&session_id, &form_csrf) {
+        return Redirect::to("/api-tokens?error=Invalid CSRF token").into_response();
+    }
+
+    use sha2::Digest;
+    let raw_token = format!("key_{}", generate_token(32));
+    let mut hasher = sha2::Sha256::new();
+    hasher.update(raw_token.as_bytes());
+    let hashed_token = hex::encode(hasher.finalize());
+
+    {
+        let conn = state.users_db.lock().unwrap();
+        let now = Utc::now().to_rfc3339();
+        let _ = conn.execute(
+            "INSERT INTO api_tokens (user_id, token_hash, created_at) VALUES (?1, ?2, ?3);",
+            rusqlite::params![user.id, hashed_token, now],
+        );
+    }
+
+    {
+        let conn_sys = state.system_db.lock().unwrap();
+        let _ = crate::db::audit_events::write_audit_event(
+            &conn_sys,
+            &user.username,
+            "API_TOKEN_CREATED",
+            "api_token",
+            "new",
+            None,
+        );
+    }
+
+    Redirect::to(&format!(
+        "/api-tokens?new_token={}&success=Token generated successfully",
+        raw_token
+    ))
+    .into_response()
+}
+
+// POST /api-tokens/revoke/:id
+pub async fn api_tokens_revoke_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    Path(token_id): Path<i64>,
+    Form(form): Form<HashMap<String, String>>,
+) -> Response {
+    let (user, session_id) = match require_user_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let form_csrf = form.get("csrf_token").cloned().unwrap_or_default();
+    if !verify_csrf(&session_id, &form_csrf) {
+        return Redirect::to("/api-tokens?error=Invalid CSRF token").into_response();
+    }
+
+    {
+        let conn = state.users_db.lock().unwrap();
+        let _ = conn.execute(
+            "DELETE FROM api_tokens WHERE id = ?1 AND user_id = ?2;",
+            [token_id, user.id],
+        );
+    }
+
+    {
+        let conn_sys = state.system_db.lock().unwrap();
+        let _ = crate::db::audit_events::write_audit_event(
+            &conn_sys,
+            &user.username,
+            "API_TOKEN_REVOKED",
+            "api_token",
+            &token_id.to_string(),
+            None,
+        );
+    }
+
+    Redirect::to("/api-tokens?success=API token revoked").into_response()
+}
@@ -0,0 +1,126 @@
+use super::*;
+
+// GET /admin/audit
+pub async fn audit_get(State(state): State<AppState>, jar: CookieJar) -> Response {
+    let (user, _) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let logs = {
+        let conn = state.system_db.lock().unwrap();
+        let events = crate::db::audit_events::list_audit_events(&conn, 100, 0, None, None)
+            .unwrap_or_default();
+        events
+            .into_iter()
+            .map(|e| {
+                let (ip, ua) = if let Some(ref m) = e.metadata {
+                    if m.starts_with("IP: ") {
+                        let parts: Vec<&str> = m.split(", UA: ").collect();
+                        let ip = parts[0]
+                            .trim_start_matches("IP: ")
+                            .trim_matches('"')
+                            .trim_matches('\'')
+                            .replace("Some(", "")
+                            .replace(")", "");
+                        let ua = if parts.len() > 1 {
+                            parts[1]
+                                .trim_matches('"')
+                                .trim_matches('\'')
+                                .replace("Some(", "")
+                                .replace(")", "")
+                        } else {
+                            "Unknown".to_string()
+                        };
+                        (Some(ip), Some(ua))
+                    } else {
+                        (None, None)
+                    }
+                } else {
+                    (None, None)
+                };
+
+                crate::models::AuditLog {
+                    id: e.id,
+                    timestamp: e.timestamp,
+                    username: e.actor,
+                    action: e.action,
+                    object_type: Some(e.object_type),
+                    object_id: Some(e.object_id),
+                    ip_address: ip,
+                    user_agent: ua,
+                }
+            })
+            .collect()
+    };
+
+    let template = crate::templates::AuditTemplate {
+        admin_username: user.username,
+        logs,
+    };
+
+    template.into_response()
+}
+
+// GET /user/audit
+pub async fn user_audit_get(State(state): State<AppState>, jar: CookieJar) -> Response {
+    let (user, _) = match require_user_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let logs = {
+        let conn = state.system_db.lock().unwrap();
+        let events =
+            crate::db::audit_events::list_audit_events(&conn, 100, 0, Some(&user.username), None)
+                .unwrap_or_default();
+        events
+            .into_iter()
+            .map(|e| {
+                let (ip, ua) = if let Some(ref m) = e.metadata {
+                    if m.starts_with("IP: ") {
+                        let parts: Vec<&str> = m.split(", UA: ").collect();
+                        let ip = parts[0]
+                            .trim_start_matches("IP: ")
+                            .trim_matches('"')
+                            .trim_matches('\'')
+                            .replace("Some(", "")
+                            .replace(")", "");
+                        let ua = if parts.len() > 1 {
+                            parts[1]
+                                .trim_matches('"')
+                                .trim_matches('\'')
+                                .replace("Some(", "")
+                                .replace(")", "")
+                        } else {
+                            "Unknown".to_string()
+                        };
+                        (Some(ip), Some(ua))
+                    } else {
+                        (None, None)
+                    }
+                } else {
+                    (None, None)
+                };
+
+                crate::models::AuditLog {
+                    id: e.id,
+                    timestamp: e.timestamp,
+                    username: e.actor,
+                    action: e.action,
+                    object_type: Some(e.object_type),
+                    object_id: Some(e.object_id),
+                    ip_address: ip,
+                    user_agent: ua,
+                }
+            })
+            .collect()
+    };
+
+    let template = crate::templates::UserAuditTemplate {
+        admin_username: user.username,
+        logs,
+    };
+
+    template.into_response()
+}
@@ -0,0 +1,517 @@
+use super::*;
+
+// GET /admin
+pub async fn admin_index(State(state): State<AppState>, jar: CookieJar) -> Response {
+    match require_auth(&state, &jar).await {
+        Ok(_) => Redirect::to("/admin/dashboard").into_response(),
+        Err(redir) => redir.into_response(),
+    }
+}
+
+// GET /admin/login
+pub async fn login_get(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    headers: axum::http::HeaderMap,
+    Query(params): Query<HashMap<String, String>>,
+) -> Response {
+    let error = params.get("error").cloned();
+    let csrf_token = generate_token(16);
+
+    let secure_flag = crate::utils::resolve_cookie_secure(state.config.cookie_secure, &headers);
+    let cookie = Cookie::build(("bzod_temp_csrf", csrf_token.clone()))
+        .path("/admin/login")
+        .secure(secure_flag)
+        .http_only(true)
+        .same_site(axum_extra::extract::cookie::SameSite::Strict)
+        .max_age(time::Duration::minutes(10))
+        .build();
+
+    let new_jar = jar.add(cookie);
+    let template = crate::templates::LoginTemplate {
+        error,
+        csrf_token,
+        action: "/admin/login".to_string(),
+        title: "Admin Login".to_string(),
+        subtitle: "Administrative Access".to_string(),
+        button_text: "Sign In".to_string(),
+    };
+    (new_jar, template).into_response()
+}
+
+#[derive(Deserialize)]
+pub struct LoginForm {
+    pub username: String,
+    pub password: String,
+    pub csrf_token: String,
+}
+
+/// Bootstrap is allowed only when the system has no real admin yet.
+pub(crate) fn is_bootstrap_allowed(
+    user_count: i64,
+    admin_count: i64,
+    active_session_count: i64,
+) -> bool {
+    user_count <= 1 && admin_count == 0 && active_session_count == 0
+}
+
+fn count_login_bootstrap_state(
+    conn: &rusqlite::Connection,
+) -> Result<(i64, i64, i64), rusqlite::Error> {
+    let u_count: i64 = conn.query_row("SELECT COUNT(*) FROM users;", [], |r| r.get(0))?;
+    let a_count: i64 = conn.query_row(
+        "SELECT COUNT(*) FROM users WHERE account_type = 'admin';",
+        [],
+        |r| r.get(0),
+    )?;
+    let now = Utc::now().to_rfc3339();
+    let s_count: i64 = conn.query_row(
+        "SELECT COUNT(*) FROM sessions WHERE expires_at > ?1;",
+        [now],
+        |r| r.get(0),
+    )?;
+    Ok((u_count, a_count, s_count))
+}
+
+/// Verify an existing admin tenant user may log into the admin UI.
+///
+/// Does not log the password. Rejection reasons are structured for observability.
+pub(crate) fn verify_admin_credentials(
+    user: &crate::models::TenantUser,
+    password: &str,
+) -> Result<(), &'static str> {
+    if user.status != "active" {
+        return Err("account_disabled");
+    }
+    if user.account_type != "admin" {
+        return Err("insufficient_privileges");
+    }
+    if !verify_password(password, &user.password_hash) {
+        return Err("invalid_credentials");
+    }
+    Ok(())
+}
+
+fn load_tenant_user_by_username(
+    conn: &rusqlite::Connection,
+    username: &str,
+) -> Result<Option<crate::models::TenantUser>, rusqlite::Error> {
+    conn.query_row(
+        "SELECT id, username, password_hash, status, created_at, last_login, account_type, organization_id, metadata
+         FROM users WHERE username = ?1;",
+        [username],
+        |row| {
+            Ok(crate::models::TenantUser {
+                id: row.get(0)?,
+                username: row.get(1)?,
+                password_hash: row.get(2)?,
+                status: row.get(3)?,
+                created_at: row.get(4)?,
+                last_login: row.get(5)?,
+                account_type: row.get(6)?,
+                organization_id: row.get(7)?,
+                metadata: row.get(8)?,
+            })
+        },
+    )
+    .optional()
+}
+
+fn tenant_user_to_admin_user(u: crate::models::TenantUser) -> User {
+    User {
+        id: u.id.to_string(),
+        username: u.username,
+        password_hash: u.password_hash,
+        created_at: u.created_at,
+    }
+}
+
+fn audit_meta(ip: &str, headers: &HeaderMap) -> String {
+    format!(
+        "IP: {:?}, UA: {:?}",
+        ip,
+        headers.get("user-agent").and_then(|h| h.to_str().ok())
+    )
+}
+
+// POST /admin/login
+pub async fn login_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    headers: HeaderMap,
+    connect_info: Option<ConnectInfo<SocketAddr>>,
+    Form(form): Form<LoginForm>,
+) -> Response {
+    let temp_csrf = jar
+        .get("bzod_temp_csrf")
+        .map(|c| c.value().to_string())
+        .unwrap_or_default();
+    if temp_csrf.is_empty() || temp_csrf != form.csrf_token {
+        return Redirect::to("/admin/login?error=Invalid CSRF token").into_response();
+    }
+
+    let ip = get_client_ip(&headers, connect_info);
+
+    let bootstrap_allowed = {
+        let conn = match state.users_db.lock() {
+            Ok(c) => c,
+            Err(_) => {
+                return Redirect::to("/admin/login?error=Internal error").into_response();
+            }
+        };
+        match count_login_bootstrap_state(&conn) {
+            Ok((u, a, s)) => is_bootstrap_allowed(u, a, s),
+            Err(e) => {
+                tracing::error!(error = %e, "login bootstrap state query failed");
+                false
+            }
+        }
+    };
+
+    let user_opt = if bootstrap_allowed
+        && form.username == state.config.admin_username
+        && verify_sha256(&form.password, &state.config.bootstrap_password_sha256)
+    {
+        // Bootstrap Phase using BOOTSTRAP_PASSWORD_SHA256
+        let hash = match hash_password(&form.password) {
+            Ok(h) => h,
+            Err(_) => {
+                return Redirect::to("/admin/login?error=Internal hashing error").into_response()
+            }
+        };
+
+        let conn = match state.users_db.lock() {
+            Ok(c) => c,
+            Err(_) => {
+                return Redirect::to("/admin/login?error=Internal error").into_response();
+            }
+        };
+        match crate::db::users::create_admin_user(&conn, &form.username, &hash) {
+            Ok(u) => {
+                if let Err(e) = state.db.init_user_databases(u.id) {
+                    tracing::error!(
+                        "Failed to init user databases during admin bootstrap: {:?}",
+                        e
+                    );
+                }
+
+                if let Ok(system_conn) = state.system_db.lock() {
+                    let metadata = audit_meta(&ip, &headers);
+                    let _ = crate::db::audit_events::write_audit_event(
+                        &system_conn,
+                        &u.username,
+                        "BOOTSTRAP_USER_PROVISIONED",
+                        "user",
+                        &u.id.to_string(),
+                        Some(&metadata),
+                    );
+                }
+
+                Some(User {
+                    id: u.id.to_string(),
+                    username: u.username,
+                    password_hash: u.password_hash,
+                    created_at: u.created_at,
+                })
+            }
+            Err(e) => {
+                tracing::error!("Failed to create admin user during bootstrap: {:?}", e);
+                None
+            }
+        }
+    } else {
+        let conn = match state.users_db.lock() {
+            Ok(c) => c,
+            Err(_) => {
+                return Redirect::to("/admin/login?error=Internal error").into_response();
+            }
+        };
+        match load_tenant_user_by_username(&conn, &form.username) {
+            Ok(Some(u)) => match verify_admin_credentials(&u, &form.password) {
+                Ok(()) => Some(tenant_user_to_admin_user(u)),
+                Err(reason) => {
+                    tracing::warn!(username = form.username, reason, "login rejected");
+                    None
+                }
+            },
+            Ok(None) => {
+                tracing::warn!(
+                    username = form.username,
+                    reason = "user_not_found",
+                    "login rejected"
+                );
+                None
+            }
+            Err(e) => {
+                tracing::error!(error = %e, "login user lookup failed");
+                None
+            }
+        }
+    };
+
+    match user_opt {
+        Some(user) => {
+            let session_token = generate_token(32);
+            let expires = (Utc::now() + chrono::Duration::days(30)).to_rfc3339();
+
+            {
+                let conn = match state.users_db.lock() {
+                    Ok(c) => c,
+                    Err(_) => {
+                        return Redirect::to("/admin/login?error=Internal error").into_response();
+                    }
+                };
+                let user_id_i64 = user.id.parse::<i64>().unwrap_or(0);
+                let now = Utc::now().to_rfc3339();
+                if let Err(e) = conn.execute(
+                    "INSERT INTO sessions (id, user_id, expires_at, created_at) VALUES (?1, ?2, ?3, ?4);",
+                    rusqlite::params![session_token, user_id_i64, expires, now],
+                ) {
+                    tracing::error!(error = %e, "failed to insert admin session");
+                    return Redirect::to("/admin/login?error=Internal error").into_response();
+                }
+
+                if let Ok(system_conn) = state.system_db.lock() {
+                    let metadata = audit_meta(&ip, &headers);
+                    let _ = crate::db::audit_events::write_audit_event(
+                        &system_conn,
+                        &user.username,
+                        "USER_LOGIN",
+                        "session",
+                        &session_token,
+                        Some(&metadata),
+                    );
+                }
+            }
+
+            let secure_flag =
+                crate::utils::resolve_cookie_secure(state.config.cookie_secure, &headers);
+            let cookie = Cookie::build(("bzod_session", session_token))
+                .path("/")
+                .secure(secure_flag)
+                .http_only(true)
+                .same_site(axum_extra::extract::cookie::SameSite::Strict)
+                .max_age(time::Duration::days(30))
+                .build();
+
+            let clear_temp = Cookie::build("bzod_temp_csrf")
+                .path("/admin/login")
+                .max_age(time::Duration::ZERO)
+                .build();
+
+            let mut response_jar = jar.clone();
+            response_jar = response_jar.add(cookie).add(clear_temp);
+
+            (response_jar, Redirect::to("/admin/dashboard")).into_response()
+        }
+        None => {
+            if let Ok(system_conn) = state.system_db.lock() {
+                let metadata = audit_meta(&ip, &headers);
+                let _ = crate::db::audit_events::write_audit_event(
+                    &system_conn,
+                    "anonymous",
+                    "LOGIN_FAILED",
+                    "login",
+                    "",
+                    Some(&metadata),
+                );
+            }
+            Redirect::to("/admin/login?error=Invalid username or password").into_response()
+        }
+    }
+}
+
+// GET /logout
+pub async fn public_logout(State(_state): State<AppState>, jar: CookieJar) -> Response {
+    let cookie = Cookie::build("bzod_user_session")
+        .path("/")
+        .max_age(time::Duration::ZERO)
+        .build();
+
+    let mut response_jar = jar.clone();
+    response_jar = response_jar.add(cookie);
+
+    (response_jar, Redirect::to("/login")).into_response()
+}
+
+// GET /login
+pub async fn public_login_get(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    headers: axum::http::HeaderMap,
+    Query(params): Query<HashMap<String, String>>,
+) -> Response {
+    let error = params.get("error").cloned();
+    let csrf_token = generate_token(16);
+
+    let secure_flag = crate::utils::resolve_cookie_secure(state.config.cookie_secure, &headers);
+    let cookie = Cookie::build(("bzod_temp_csrf", csrf_token.clone()))
+        .path("/login")
+        .secure(secure_flag)
+        .http_only(true)
+        .same_site(axum_extra::extract::cookie::SameSite::Strict)
+        .max_age(time::Duration::minutes(10))
+        .build();
+
+    let new_jar = jar.add(cookie);
+    let template = crate::templates::LoginTemplate {
+        error,
+        csrf_token,
+        action: "/login".to_string(),
+        title: "User Login".to_string(),
+        subtitle: "Standard account access".to_string(),
+        button_text: "Sign In".to_string(),
+    };
+    (new_jar, template).into_response()
+}
+
+// POST /login
+pub async fn public_login_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    headers: HeaderMap,
+    connect_info: Option<ConnectInfo<SocketAddr>>,
+    Form(form): Form<LoginForm>,
+) -> Response {
+    let temp_csrf = jar
+        .get("bzod_temp_csrf")
+        .map(|c| c.value().to_string())
+        .unwrap_or_default();
+    if temp_csrf.is_empty() || temp_csrf != form.csrf_token {
+        return Redirect::to("/login?error=Invalid CSRF token").into_response();
+    }
+
+    let ip = get_client_ip(&headers, connect_info);
+
+    let user_opt: Option<crate::models::TenantUser> = {
+        let conn = state.users_db.lock().unwrap();
+        let user_res: Result<Option<crate::models::TenantUser>, rusqlite::Error> = conn.query_row(
+            "SELECT id, username, password_hash, status, created_at, last_login, account_type, organization_id, metadata \
+             FROM users WHERE username = ?1;",
+            [&form.username],
+            |row| {
+                Ok(crate::models::TenantUser {
+                    id: row.get(0)?,
+                    username: row.get(1)?,
+                    password_hash: row.get(2)?,
+                    status: row.get(3)?,
+                    created_at: row.get(4)?,
+                    last_login: row.get(5)?,
+                    account_type: row.get(6)?,
+                    organization_id: row.get(7)?,
+                    metadata: row.get(8)?,
+                })
+            }
+        ).optional();
+
+        match user_res {
+            Ok(Some(u)) => {
+                if u.status != "active" {
+                    None
+                } else if verify_password(&form.password, &u.password_hash) {
+                    Some(u)
+                } else {
+                    None
+                }
+            }
+            _ => None,
+        }
+    };
+
+    match user_opt {
+        Some(user) => {
+            let session_token = generate_token(32);
+            let expires = (Utc::now() + chrono::Duration::days(30)).to_rfc3339();
+
+            {
+                let conn = state.users_db.lock().unwrap();
+                let _ =
+                    crate::db::users::create_user_session(&conn, &session_token, user.id, &expires);
+
+                let system_conn = state.system_db.lock().unwrap();
+                let metadata = format!(
+                    "IP: {:?}, UA: {:?}",
+                    ip,
+                    headers.get("user-agent").and_then(|h| h.to_str().ok())
+                );
+                let _ = crate::db::audit_events::write_audit_event(
+                    &system_conn,
+                    &user.username,
+                    "USER_LOGIN",
+                    "session",
+                    &session_token,
+                    Some(&metadata),
+                );
+            }
+
+            let secure_flag =
+                crate::utils::resolve_cookie_secure(state.config.cookie_secure, &headers);
+            let cookie = Cookie::build(("bzod_user_session", session_token))
+                .path("/")
+                .secure(secure_flag)
+                .http_only(true)
+                .same_site(axum_extra::extract::cookie::SameSite::Strict)
+                .max_age(time::Duration::days(30))
+                .build();
+
+            let clear_temp = Cookie::build("bzod_temp_csrf")
+                .path("/login")
+                .max_age(time::Duration::ZERO)
+                .build();
+
+            let mut response_jar = jar.clone();
+            response_jar = response_jar.add(cookie).add(clear_temp);
+
+            (response_jar, Redirect::to("/user/dashboard")).into_response()
+        }
+        None => {
+            let system_conn = state.system_db.lock().unwrap();
+            let metadata = format!(
+                "IP: {:?}, UA: {:?}",
+                ip,
+                headers.get("user-agent").and_then(|h| h.to_str().ok())
+            );
+            let _ = crate::db::audit_events::write_audit_event(
+                &system_conn,
+                "anonymous",
+                "LOGIN_FAILED",
+                "login",
+                "",
+                Some(&metadata),
+            );
+            Redirect::to("/login?error=Invalid username or password").into_response()
+        }
+    }
+}
+
+// GET /admin/logout
+pub async fn logout(State(state): State<AppState>, jar: CookieJar) -> Response {
+    if let Ok((_, session_id)) = require_auth(&state, &jar).await {
+        let conn = state.users_db.lock().unwrap();
+        let _ = conn.execute("DELETE FROM sessions WHERE id = ?1;", [&session_id]);
+    }
+
+    let cookie = Cookie::build("bzod_session")
+        .path("/")
+        .max_age(time::Duration::ZERO)
+        .build();
+
+    let mut response_jar = jar.clone();
+    response_jar = response_jar.add(cookie);
+
+    (response_jar, Redirect::to("/admin/login")).into_response()
+}
+
+#[cfg(test)]
+mod login_helpers_tests {
+    use super::is_bootstrap_allowed;
+
+    #[test]
+    fn bootstrap_only_when_no_admin() {
+        assert!(is_bootstrap_allowed(0, 0, 0));
+        assert!(is_bootstrap_allowed(1, 0, 0));
+        assert!(!is_bootstrap_allowed(2, 0, 0));
+        assert!(!is_bootstrap_allowed(1, 1, 0));
+        assert!(!is_bootstrap_allowed(1, 0, 1));
+    }
+}
@@ -0,0 +1,300 @@
+use super::*;
+
+#[derive(Clone, Debug, serde::Serialize, serde::Deserialize)]
+pub struct BackupFileRow {
+    pub filename: String,
+    pub size_str: String,
+    pub created_str: String,
+}
+
+#[derive(Clone, Debug, serde::Serialize, serde::Deserialize)]
+pub struct BackupHistoryRow {
+    pub id: String,
+    pub backup_path: String,
+    pub status: String,
+    pub created_at: String,
+    pub size_bytes: i64,
+    pub error_message: Option<String>,
+}
+
+// GET /admin/backups
+pub async fn backups_get(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    Query(params): Query<HashMap<String, String>>,
+) -> Response {
+    let (user, session_id) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let mut files = vec![];
+    if let Ok(dir_entries) = std::fs::read_dir(&state.config.backup_dir) {
+        for entry in dir_entries.flatten() {
+            let path = entry.path();
+            if path.is_file() {
+                if let Some(filename) = path
+                    .file_name()
+                    .and_then(|n| n.to_str())
+                    .map(|s| s.to_string())
+                {
+                    if filename.ends_with(".tar.gz") {
+                        let meta = entry.metadata().unwrap();
+                        let size_str = format_size(meta.len());
+                        let created_str = meta
+                            .created()
+                            .ok()
+                            .map(|c| {
+                                let datetime: chrono::DateTime<chrono::Utc> = c.into();
+                                datetime.format("%Y-%m-%d %H:%M:%S").to_string()
+                            })
+                            .unwrap_or_else(|| "-".to_string());
+                        files.push(BackupFileRow {
+                            filename,
+                            size_str,
+                            created_str,
+                        });
+                    }
+                }
+            }
+        }
+    }
+    files.sort_by(|a, b| b.filename.cmp(&a.filename));
+
+    let history = {
+        let conn = state.system_db.lock().unwrap();
+        let mut stmt = conn.prepare("SELECT id, backup_path, status, created_at, size_bytes, error_message FROM backup_history ORDER BY created_at DESC LIMIT 30;").unwrap();
+        let rows = stmt
+            .query_map([], |row| {
+                Ok(BackupHistoryRow {
+                    id: row.get(0)?,
+                    backup_path: row.get(1)?,
+                    status: row.get(2)?,
+                    created_at: row.get(3)?,
+                    size_bytes: row.get(4)?,
+                    error_message: row.get(5)?,
+                })
+            })
+            .unwrap();
+        rows.filter_map(|r| r.ok()).collect()
+    };
+
+    let csrf_token = generate_csrf_token(&session_id);
+
+    let template = crate::templates::BackupsTemplate {
+        admin_username: user.username,
+        files,
+        history,
+        csrf_token,
+        success: params.get("success").cloned(),
+        error: params.get("error").cloned(),
+    };
+
+    template.into_response()
+}
+
+// POST /admin/backups/create
+pub async fn backups_create_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    Form(form): Form<HashMap<String, String>>,
+) -> Response {
+    let (_user, session_id) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let form_csrf = form.get("csrf_token").cloned().unwrap_or_default();
+    if !verify_csrf(&session_id, &form_csrf) {
+        return Redirect::to("/admin/backups?error=Invalid CSRF token").into_response();
+    }
+
+    match crate::jobs::backup::perform_backup(&state.db, &state.config).await {
+        Ok(path) => {
+            let filename = std::path::Path::new(&path)
+                .file_name()
+                .and_then(|n| n.to_str())
+                .unwrap_or("backup.tar.gz");
+            Redirect::to(&format!(
+                "/admin/backups?success=Backup created successfully: {}",
+                filename
+            ))
+            .into_response()
+        }
+        Err(e) => Redirect::to(&format!(
+            "/admin/backups?error=Failed to generate backup: {}",
+            e
+        ))
+        .into_response(),
+    }
+}
+
+// GET /admin/backups/download/:filename
+pub async fn backups_download_get(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    Path(filename): Path<String>,
+) -> Response {
+    if require_auth(&state, &jar).await.is_err() {
+        return StatusCode::UNAUTHORIZED.into_response();
+    }
+
+    if filename.contains('/') || filename.contains('\\') || filename.contains("..") {
+        return StatusCode::BAD_REQUEST.into_response();
+    }
+
+    let backup_path = state.config.backup_dir.join(&filename);
+    if !backup_path.exists() {
+        return StatusCode::NOT_FOUND.into_response();
+    }
+
+    match std::fs::read(&backup_path) {
+        Ok(bytes) => {
+            let body = axum::body::Body::from(bytes);
+            Response::builder()
+                .header("content-type", "application/octet-stream")
+                .header(
+                    "content-disposition",
+                    format!("attachment; filename=\"{}\"", filename),
+                )
+                .body(body)
+                .unwrap_or_else(|_| StatusCode::INTERNAL_SERVER_ERROR.into_response())
+        }
+        Err(_) => StatusCode::INTERNAL_SERVER_ERROR.into_response(),
+    }
+}
+
+// POST /admin/backups/delete/:filename
+pub async fn backups_delete_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    Path(filename): Path<String>,
+    Form(form): Form<HashMap<String, String>>,
+) -> Response {
+    let (user, session_id) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let form_csrf = form.get("csrf_token").cloned().unwrap_or_default();
+    if !verify_csrf(&session_id, &form_csrf) {
+        return Redirect::to("/admin/backups?error=Invalid CSRF token").into_response();
+    }
+
+    if filename.contains('/') || filename.contains('\\') || filename.contains("..") {
+        return Redirect::to("/admin/backups?error=Invalid filename").into_response();
+    }
+
+    let backup_path = state.config.backup_dir.join(&filename);
+    if !backup_path.exists() {
+        return Redirect::to("/admin/backups?error=Backup file not found").into_response();
+    }
+
+    match std::fs::remove_file(&backup_path) {
+        Ok(_) => {
+            let system_conn = state.system_db.lock().unwrap();
+            let _ = crate::db::audit_events::write_audit_event(
+                &system_conn,
+                &user.username,
+                "BACKUP_DELETE",
+                "backup",
+                &filename,
+                None,
+            );
+            Redirect::to("/admin/backups?success=Backup archive deleted").into_response()
+        }
+        Err(e) => Redirect::to(&format!(
+            "/admin/backups?error=Failed to delete backup file: {}",
+            e
+        ))
+        .into_response(),
+    }
+}
+
+// POST /admin/backups/restore
+pub async fn backups_restore_post(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    mut multipart: axum::extract::Multipart,
+) -> Response {
+    let (user, session_id) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let mut backup_bytes = vec![];
+    let mut confirm_text = String::new();
+    let mut csrf_token = String::new();
+
+    while let Ok(Some(field)) = multipart.next_field().await {
+        let name = field.name().unwrap_or_default().to_string();
+        if name == "backup_file" {
+            if let Ok(bytes) = field.bytes().await {
+                backup_bytes = bytes.to_vec();
+            }
+        } else if name == "confirm_text" {
+            if let Ok(text) = field.text().await {
+                confirm_text = text.trim().to_string();
+            }
+        } else if name == "csrf_token" {
+            if let Ok(text) = field.text().await {
+                csrf_token = text.trim().to_string();
+            }
+        }
+    }
+
+    if !verify_csrf(&session_id, &csrf_token) {
+        return Redirect::to("/admin/backups?error=Invalid CSRF token").into_response();
+    }
+
+    if confirm_text != "RESTORE" {
+        return Redirect::to(
+            "/admin/backups?error=Confirmation text mismatch. Please type RESTORE.",
+        )
+        .into_response();
+    }
+
+    if backup_bytes.is_empty() {
+        return Redirect::to("/admin/backups?error=Backup file is empty or missing.")
+            .into_response();
+    }
+
+    let temp_file_path = state.config.data_dir.join("temp-restore-upload.tar.gz");
+    if let Err(e) = std::fs::write(&temp_file_path, &backup_bytes) {
+        return Redirect::to(&format!(
+            "/admin/backups?error=Failed to save uploaded file: {}",
+            e
+        ))
+        .into_response();
+    }
+
+    match crate::cli::restore::run(
+        temp_file_path.to_string_lossy().to_string(),
+        None,
+        state.config.clone(),
+    )
+    .await
+    {
+        Ok(_) => {
+            let _ = std::fs::remove_file(&temp_file_path);
+            let conn = state.users_db.lock().unwrap();
+            let _ = conn.execute("DELETE FROM sessions;", []);
+
+            let system_conn = state.system_db.lock().unwrap();
+            let _ = crate::db::audit_events::write_audit_event(
+                &system_conn,
+                &user.username,
+                "RESTORE_EXECUTION",
+                "backup",
+                "upload",
+                None,
+            );
+            Redirect::to("/admin/login?success=Restore successful. Please log in again.")
+                .into_response()
+        }
+        Err(e) => {
+            let _ = std::fs::remove_file(&temp_file_path);
+            Redirect::to(&format!("/admin/backups?error=Restore failed: {}", e)).into_response()
+        }
+    }
+}
@@ -0,0 +1,170 @@
+use super::*;
+
+// GET /user/dashboard
+pub async fn user_dashboard_get(State(state): State<AppState>, jar: CookieJar) -> Response {
+    let (user, _session_id) = match require_user_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let user_dbs = match state.get_user_dbs(user.id) {
+        Ok(dbs) => dbs,
+        Err(_) => return (StatusCode::INTERNAL_SERVER_ERROR, "Database error").into_response(),
+    };
+
+    let (total_urls, active_links, dead_links) = {
+        let conn = user_dbs.content.lock().unwrap();
+        get_url_counts(&conn).unwrap_or((0, 0, 0))
+    };
+
+    let total_pages = {
+        let conn = user_dbs.content.lock().unwrap();
+        get_landing_page_count(&conn).unwrap_or(0)
+    };
+
+    let total_clicks = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        get_total_clicks(&conn).unwrap_or(0)
+    };
+
+    let clicks_data = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        get_clicks_trend(&conn, "url", "all", 30)
+            .or_else(|_| get_clicks_trend_raw(&conn, "url", "all", 30))
+            .unwrap_or_default()
+    };
+
+    let mut trend_map = std::collections::BTreeMap::new();
+    for i in (0..30).rev() {
+        let date_str = (Utc::now() - chrono::Duration::days(i))
+            .format("%Y-%m-%d")
+            .to_string();
+        trend_map.insert(date_str, 0i64);
+    }
+    for (d, c) in clicks_data {
+        trend_map.insert(d, c);
+    }
+    let formatted_trend: Vec<(String, i64)> = trend_map.into_iter().collect();
+    let traffic_chart = generate_line_chart(&formatted_trend);
+
+    let countries_data = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        get_metric_rankings(&conn, "url", "all", "country", 5)
+            .or_else(|_| get_metric_rankings_raw(&conn, "url", "all", "country", 5))
+            .unwrap_or_default()
+    };
+    let countries_chart = generate_bar_chart(&countries_data);
+
+    let referrers_data = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        get_metric_rankings(&conn, "url", "all", "referrer", 5)
+            .or_else(|_| get_metric_rankings_raw(&conn, "url", "all", "referrer", 5))
+            .unwrap_or_default()
+    };
+    let referrers_chart = generate_bar_chart(&referrers_data);
+
+    let browsers_data = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        get_metric_rankings(&conn, "url", "all", "browser", 5)
+            .or_else(|_| get_metric_rankings_raw(&conn, "url", "all", "browser", 5))
+            .unwrap_or_default()
+    };
+    let browsers_chart = generate_bar_chart(&browsers_data);
+
+    let template = crate::templates::UserDashboardTemplate {
+        admin_username: user.username,
+        total_urls,
+        total_pages,
+        total_clicks,
+        active_links,
+        dead_links,
+        traffic_chart,
+        countries_chart,
+        browsers_chart,
+        referrers_chart,
+    };
+
+    template.into_response()
+}
+
+// GET /admin/dashboard
+pub async fn dashboard_get(State(state): State<AppState>, jar: CookieJar) -> Response {
+    let (user, _) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let (total_urls, active_links, dead_links) = {
+        let conn = state.content_db.lock().unwrap();
+        get_url_counts(&conn).unwrap_or((0, 0, 0))
+    };
+
+    let total_pages = {
+        let conn = state.content_db.lock().unwrap();
+        get_landing_page_count(&conn).unwrap_or(0)
+    };
+
+    let total_clicks = {
+        let conn = state.analytics_db.lock().unwrap();
+        get_total_clicks(&conn).unwrap_or(0)
+    };
+
+    let clicks_data = {
+        let conn = state.analytics_db.lock().unwrap();
+        get_clicks_trend(&conn, "url", "all", 30)
+            .or_else(|_| get_clicks_trend_raw(&conn, "url", "all", 30))
+            .unwrap_or_default()
+    };
+
+    let mut trend_map = std::collections::BTreeMap::new();
+    for i in (0..30).rev() {
+        let date_str = (Utc::now() - chrono::Duration::days(i))
+            .format("%Y-%m-%d")
+            .to_string();
+        trend_map.insert(date_str, 0i64);
+    }
+    for (d, c) in clicks_data {
+        trend_map.insert(d, c);
+    }
+    let formatted_trend: Vec<(String, i64)> = trend_map.into_iter().collect();
+    let traffic_chart = generate_line_chart(&formatted_trend);
+
+    let countries_data = {
+        let conn = state.analytics_db.lock().unwrap();
+        get_metric_rankings(&conn, "url", "all", "country", 5)
+            .or_else(|_| get_metric_rankings_raw(&conn, "url", "all", "country", 5))
+            .unwrap_or_default()
+    };
+    let countries_chart = generate_bar_chart(&countries_data);
+
+    let referrers_data = {
+        let conn = state.analytics_db.lock().unwrap();
+        get_metric_rankings(&conn, "url", "all", "referrer", 5)
+            .or_else(|_| get_metric_rankings_raw(&conn, "url", "all", "referrer", 5))
+            .unwrap_or_default()
+    };
+    let referrers_chart = generate_bar_chart(&referrers_data);
+
+    let browsers_data = {
+        let conn = state.analytics_db.lock().unwrap();
+        get_metric_rankings(&conn, "url", "all", "browser", 5)
+            .or_else(|_| get_metric_rankings_raw(&conn, "url", "all", "browser", 5))
+            .unwrap_or_default()
+    };
+    let browsers_chart = generate_bar_chart(&browsers_data);
+
+    let template = crate::templates::DashboardTemplate {
+        admin_username: user.username,
+        total_urls,
+        total_pages,
+        total_clicks,
+        active_links,
+        dead_links,
+        traffic_chart,
+        countries_chart,
+        browsers_chart,
+        referrers_chart,
+    };
+
+    template.into_response()
+}
@@ -0,0 +1,169 @@
+use super::*;
+
+#[derive(Clone, Debug, serde::Serialize, serde::Deserialize)]
+pub struct JobHistoryRow {
+    pub id: String,
+    pub job_name: String,
+    pub status: String,
+    pub started_at: String,
+    pub finished_at: Option<String>,
+    pub error_message: Option<String>,
+}
+
+#[derive(Clone, Debug, serde::Serialize, serde::Deserialize)]
+pub struct HealthCheckRow {
+    pub id: String,
+    pub object_type: String,
+    pub object_id: String,
+    pub checked_at: String,
+    pub status_code: Option<i64>,
+    pub error_message: Option<String>,
+    pub is_healthy: i64,
+}
+
+// GET /admin/health
+pub async fn health_get(
+    State(state): State<AppState>,
+    jar: CookieJar,
+    Query(params): Query<HashMap<String, String>>,
+) -> Response {
+    let (user, session_id) = match require_auth(&state, &jar).await {
+        Ok(u) => u,
+        Err(redir) => return redir.into_response(),
+    };
+
+    let mut db_reports = vec![];
+    if let Ok(r) =
+        crate::db::sqlite::collect_health_report(&state.admin_db.lock().unwrap(), "admin")
+    {
+        db_reports.push(r);
+    }
+    if let Ok(r) =
+        crate::db::sqlite::collect_health_report(&state.system_db.lock().unwrap(), "system")
+    {
+        db_reports.push(r);
+    }
+    if let Ok(r) =
+        crate::db::sqlite::collect_health_report(&state.users_db.lock().unwrap(), "users")
+    {
+        db_reports.push(r);
+    }
+
+    let system_db_path = state.config.data_dir.join("admin").join("system.db");
+    let users_db_path = state.config.data_dir.join("admin").join("users.db");
+    let admin_db_path = state.config.data_dir.join("admin").join("admin.db");
+
+    let system_db_size = format_size(
+        std::fs::metadata(&system_db_path)
+            .map(|m| m.len())
+            .unwrap_or(0),
+    );
+    let users_db_size = format_size(
+        std::fs::metadata(&users_db_path)
+            .map(|m| m.len())
+            .unwrap_or(0),
+    );
+    let admin_db_size = format_size(
+        std::fs::metadata(&admin_db_path)
+            .map(|m| m.len())
+            .unwrap_or(0),
+    );
+
+    let users_dir = state.config.data_dir.join("users");
+    let tenants_db_size = format_size(get_dir_size(&users_dir).unwrap_or(0));
+    let total_data_size = format_size(get_dir_size(&state.config.data_dir).unwrap_or(0));
+
+    let job_history = {
+        let conn = state.system_db.lock().unwrap();
+        let mut stmt = conn.prepare("SELECT id, job_name, status, started_at, finished_at, error_message FROM job_history ORDER BY started_at DESC LIMIT 20;").unwrap();
+        let rows = stmt
+            .query_map([], |row| {
+                Ok(JobHistoryRow {
+                    id: row.get(0)?,
+                    job_name: row.get(1)?,
+                    status: row.get(2)?,
+                    started_at: row.get(3)?,
+                    finished_at: row.get(4)?,
+                    error_message: row.get(5)?,
+                })
+            })
+            .unwrap();
+        rows.filter_map(|r| r.ok()).collect()
+    };
+
+    let health_checks = {
+        let conn = state.system_db.lock().unwrap();
+        let mut stmt = conn.prepare("SELECT id, object_type, object_id, checked_at, status_code, error_message, is_healthy FROM health_checks ORDER BY checked_at DESC LIMIT 20;").unwrap();
+        let rows = stmt
+            .query_map([], |row| {
+                Ok(HealthCheckRow {
+                    id: row.get(0)?,
+                    object_type: row.get(1)?,
+                    object_id: row.get(2)?,
+                    checked_at: row.get(3)?,
+                    status_code: row.get(4)?,
+                    error_message: row.get(5)?,
+                    is_healthy: row.get(6)?,
+                })
+            })
+            .unwrap();
+        rows.filter_map(|r| r.ok()).collect()
+    };
+
+    let csrf_token = generate_csrf_token(&session_id);
+
+    let (registry_errors, registry_warnings) = {
+        let system_conn = state.system_db.lock().unwrap();
+        let users_conn = state.users_db.lock().unwrap();
+        match crate::services::registry_validator::RegistryValidator::scan(
+            &system_conn,
+            &users_conn,
+            &state.config.data_dir,
+            None,
+        ) {
+            Ok(issues) => {
+                let mut errors = Vec::new();
+                let mut warnings = Vec::new();
+                for issue in issues {
+                    use crate::services::registry_validator::RegistryIssueType;
+                    match issue.issue_type {
+                        RegistryIssueType::StaleReservation
+                        | RegistryIssueType::TenantAdminHasIsolatedContent => {
+                            warnings.push(format!(
+                                "Warning for slug {}: {}",
+                                issue.slug, issue.description
+                            ));
+                        }
+                        _ => {
+                            errors.push(format!(
+                                "Error for slug {}: {}",
+                                issue.slug, issue.description
+                            ));
+                        }
+                    }
+                }
+                (errors, warnings)
+            }
+            Err(e) => (vec![format!("Failed to run registry scan: {}", e)], vec![]),
+        }
+    };
+
+    let template = crate::templates::HealthTemplate {
+        admin_username: user.username,
+        db_reports,
+        total_data_size,
+        system_db_size,
+        users_db_size,
+        admin_db_size,
+        tenants_db_size,
+        job_history,
+        health_checks,
+        registry_errors,
+        registry_warnings,
+        csrf_token,
+        success: params.get("success").cloned(),
+        error: params.get("error").cloned(),
+    };
+
+    template.into_response()
+}
@@ -0,0 +1,849 @@
+//! Admin web UI handlers (feature-split modules).
+//!
+//! Handlers are organized by domain; shared auth, audit, export, and helpers
+//! live in this module root so child modules can access them via `super`.
+
+use crate::auth::{
+    authenticate_admin_session, authenticate_user_session, generate_csrf_token, generate_token,
+    hash_password, verify_csrf, verify_password, verify_sha256,
+};
+use crate::charts::{generate_bar_chart, generate_line_chart};
+use crate::db::admin::{
+    create_api_key, delete_api_key, get_config, get_user_count, list_api_keys, set_config,
+    write_audit_log as write_audit_log_legacy,
+};
+use crate::db::analytics::{
+    clean_referrer, get_clicks_trend, get_clicks_trend_raw, get_metric_rankings,
+    get_metric_rankings_raw, get_monthly_clicks_trend, get_target_unique_visitors,
+    get_target_visit_total_filtered, get_target_visits_all_in_memory, get_target_visits_paginated,
+    get_total_clicks, get_visits_schema_columns, parse_ua,
+};
+use crate::db::content::{
+    create_landing_page, delete_landing_page, delete_url, get_landing_page_by_id,
+    get_landing_page_count, get_url_by_id, get_url_count_by_tag, get_url_counts,
+    list_landing_pages, list_urls,
+};
+use crate::models::User;
+use crate::state::AppState;
+use crate::utils::{get_client_ip, get_db_file_info, get_memory_usage};
+use axum::{
+    extract::{ConnectInfo, Path, Query, State},
+    http::{HeaderMap, StatusCode},
+    response::{IntoResponse, Redirect, Response},
+    Form,
+};
+use axum_extra::extract::cookie::Cookie;
+use axum_extra::extract::CookieJar;
+use chrono::Utc;
+use flate2::read::GzDecoder;
+use flate2::write::GzEncoder;
+use flate2::Compression;
+use rusqlite::{params, OptionalExtension};
+use serde::Deserialize;
+use std::collections::HashMap;
+use std::fs::File;
+use std::net::SocketAddr;
+use tar::Builder;
+use uuid::Uuid;
+
+// --- Shared constants, auth, audit, and export helpers ---
+
+#[allow(clippy::too_many_arguments)]
+pub(crate) fn write_audit_log(
+    conn: &rusqlite::Connection,
+    state: &AppState,
+    username: &str,
+    action: &str,
+    object_type: Option<&str>,
+    object_id: Option<&str>,
+    ip_address: Option<&str>,
+    user_agent: Option<&str>,
+) -> rusqlite::Result<crate::models::AuditLog> {
+    let res = write_audit_log_legacy(
+        conn,
+        username,
+        action,
+        object_type,
+        object_id,
+        ip_address,
+        user_agent,
+    );
+
+    // Also write to unified audit events in system.db
+    let system_conn = state.system_db.lock().unwrap();
+    let metadata = format!("IP: {:?}, UA: {:?}", ip_address, user_agent);
+    let _ = crate::db::audit_events::write_audit_event(
+        &system_conn,
+        username,
+        action,
+        object_type.unwrap_or(""),
+        object_id.unwrap_or(""),
+        Some(&metadata),
+    );
+
+    res
+}
+pub(crate) const PAGE_SIZE: usize = 25;
+pub(crate) const ANALYTICS_PAGE_SIZE: usize = 50;
+pub(crate) const MAX_JSON_EXPORT_ROWS: usize = 50_000;
+// Helper: Verify admin session and return user or redirect to login
+pub(crate) async fn require_auth(
+    state: &AppState,
+    jar: &CookieJar,
+) -> Result<(User, String), Redirect> {
+    let conn = match state.users_db.lock() {
+        Ok(c) => c,
+        Err(_) => return Err(Redirect::to("/admin/login")),
+    };
+    match authenticate_admin_session(&conn, jar) {
+        Ok(Some((user, session_id))) => Ok((user, session_id)),
+        _ => Err(Redirect::to("/admin/login")),
+    }
+}
+// Helper: Verify tenant user session and return user or redirect to login
+pub(crate) async fn require_user_auth(
+    state: &AppState,
+    jar: &CookieJar,
+) -> Result<(crate::models::TenantUser, String), Redirect> {
+    let conn = match state.users_db.lock() {
+        Ok(c) => c,
+        Err(_) => return Err(Redirect::to("/login")),
+    };
+    match authenticate_user_session(&conn, jar) {
+        Ok(Some((user, session_id))) => Ok((user, session_id)),
+        _ => Err(Redirect::to("/login")),
+    }
+}
+#[derive(Deserialize)]
+pub struct AnalyticsQuery {
+    pub analytics_page: Option<usize>,
+    pub date_from: Option<String>,
+    pub date_to: Option<String>,
+}
+pub(crate) fn validate_date_filters(
+    date_from: Option<&str>,
+    date_to: Option<&str>,
+) -> Result<(Option<String>, Option<String>), StatusCode> {
+    let from_parsed = match date_from {
+        Some(df) if !df.is_empty() => match chrono::NaiveDate::parse_from_str(df, "%Y-%m-%d") {
+            Ok(d) => Some(d),
+            Err(_) => return Err(StatusCode::BAD_REQUEST),
+        },
+        _ => None,
+    };
+    let to_parsed = match date_to {
+        Some(dt) if !dt.is_empty() => match chrono::NaiveDate::parse_from_str(dt, "%Y-%m-%d") {
+            Ok(d) => Some(d),
+            Err(_) => return Err(StatusCode::BAD_REQUEST),
+        },
+        _ => None,
+    };
+    if let (Some(f), Some(t)) = (from_parsed, to_parsed) {
+        if f > t {
+            return Err(StatusCode::BAD_REQUEST);
+        }
+    }
+    Ok((
+        from_parsed.map(|d| d.format("%Y-%m-%d").to_string()),
+        to_parsed.map(|d| d.format("%Y-%m-%d").to_string()),
+    ))
+}
+pub(crate) fn escape_csv_field(field: &str) -> String {
+    let needs_escaping =
+        field.contains(',') || field.contains('"') || field.contains('\n') || field.contains('\r');
+    if needs_escaping {
+        let escaped = field.replace('"', "\"\"");
+        format!("\"{}\"", escaped)
+    } else {
+        field.to_string()
+    }
+}
+pub(crate) struct DbExportStream {
+    receiver: tokio::sync::mpsc::Receiver<Result<axum::body::Bytes, std::convert::Infallible>>,
+}
+
+impl futures_util::stream::Stream for DbExportStream {
+    type Item = Result<axum::body::Bytes, std::convert::Infallible>;
+
+    fn poll_next(
+        mut self: std::pin::Pin<&mut Self>,
+        cx: &mut std::task::Context<'_>,
+    ) -> std::task::Poll<Option<Self::Item>> {
+        self.receiver.poll_recv(cx)
+    }
+}
+
+pub(crate) async fn perform_csv_export(
+    state: AppState,
+    target_type: &'static str,
+    id: String,
+    date_from: Option<String>,
+    date_to: Option<String>,
+) -> Response {
+    let (clean_date_from, clean_date_to) =
+        match validate_date_filters(date_from.as_deref(), date_to.as_deref()) {
+            Ok(res) => res,
+            Err(status) => return status.into_response(),
+        };
+
+    let target_exists = {
+        let conn = state.content_db.lock().unwrap();
+        if target_type == "url" {
+            get_url_by_id(&conn, &id)
+                .map(|u| u.is_some())
+                .unwrap_or(false)
+        } else {
+            get_landing_page_by_id(&conn, &id)
+                .map(|p| p.is_some())
+                .unwrap_or(false)
+        }
+    };
+    if !target_exists {
+        return (StatusCode::NOT_FOUND, "Target not found").into_response();
+    }
+
+    let count = {
+        let conn = state.analytics_db.lock().unwrap();
+        get_target_visit_total_filtered(
+            &conn,
+            target_type,
+            &id,
+            clean_date_from.as_deref(),
+            clean_date_to.as_deref(),
+        )
+        .unwrap_or(0)
+    };
+
+    let (has_utm_source, has_utm_campaign) = {
+        let conn = state.analytics_db.lock().unwrap();
+        let cols = get_visits_schema_columns(&conn).unwrap_or_default();
+        (cols.contains("utm_source"), cols.contains("utm_campaign"))
+    };
+
+    let (tx, rx) =
+        tokio::sync::mpsc::channel::<Result<axum::body::Bytes, std::convert::Infallible>>(32);
+    let analytics_db = state.analytics_db.clone();
+    let target_id = id.clone();
+
+    tokio::task::spawn_blocking(move || {
+        let conn = analytics_db.lock().unwrap();
+
+        let mut header = "Timestamp,IP Address,Country,Referrer,Browser,User-Agent".to_string();
+        if has_utm_source {
+            header.push_str(",UTM Source");
+        }
+        if has_utm_campaign {
+            header.push_str(",UTM Campaign");
+        }
+        header.push('\n');
+
+        if tx
+            .blocking_send(Ok(axum::body::Bytes::from(header)))
+            .is_err()
+        {
+            return;
+        }
+
+        let select_fields = if has_utm_source && has_utm_campaign {
+            "timestamp, ip_address, country, referer, user_agent, utm_source, utm_campaign"
+        } else if has_utm_source {
+            "timestamp, ip_address, country, referer, user_agent, utm_source"
+        } else if has_utm_campaign {
+            "timestamp, ip_address, country, referer, user_agent, utm_campaign"
+        } else {
+            "timestamp, ip_address, country, referer, user_agent"
+        };
+
+        let mut sql = format!(
+            "SELECT {} FROM visits WHERE target_type = ?1 AND target_id = ?2",
+            select_fields
+        );
+        let mut params: Vec<Box<dyn rusqlite::ToSql>> =
+            vec![Box::new(target_type.to_string()), Box::new(target_id)];
+
+        if let Some(df) = clean_date_from.as_deref() {
+            sql.push_str(&format!(" AND timestamp >= ?{}", params.len() + 1));
+            params.push(Box::new(format!("{}T00:00:00Z", df)));
+        }
+
+        if let Some(dt) = clean_date_to.as_deref() {
+            if let Ok(parsed_date) = chrono::NaiveDate::parse_from_str(dt, "%Y-%m-%d") {
+                let next_day = parsed_date + chrono::Duration::days(1);
+                sql.push_str(&format!(" AND timestamp < ?{}", params.len() + 1));
+                params.push(Box::new(format!(
+                    "{}T00:00:00Z",
+                    next_day.format("%Y-%m-%d")
+                )));
+            }
+        }
+
+        sql.push_str(" ORDER BY timestamp DESC, id DESC");
+
+        let mut stmt = match conn.prepare(&sql) {
+            Ok(s) => s,
+            Err(_) => return,
+        };
+
+        let param_refs: Vec<&dyn rusqlite::ToSql> = params.iter().map(|p| p.as_ref()).collect();
+        let mut rows = match stmt.query(rusqlite::params_from_iter(param_refs)) {
+            Ok(r) => r,
+            Err(_) => return,
+        };
+
+        let mut csv_buffer = String::new();
+
+        while let Ok(Some(row)) = rows.next() {
+            let timestamp: String = row.get(0).unwrap_or_default();
+            let ip_address: String = row.get(1).unwrap_or_default();
+            let country: String = row.get(2).unwrap_or_default();
+            let referer: String = row.get(3).unwrap_or_default();
+            let user_agent: String = row.get(4).unwrap_or_default();
+
+            let (browser, _, _) = parse_ua(&user_agent);
+            let referrer = clean_referrer(&referer);
+            let country_display = if country.is_empty() {
+                "Unknown".to_string()
+            } else {
+                country
+            };
+
+            let mut line = format!(
+                "{},{},{},{},{},{}",
+                escape_csv_field(&timestamp),
+                escape_csv_field(&ip_address),
+                escape_csv_field(&country_display),
+                escape_csv_field(&referrer),
+                escape_csv_field(&browser),
+                escape_csv_field(&user_agent)
+            );
+
+            let mut col_idx = 5;
+            if has_utm_source {
+                let utm_src: String = row.get(col_idx).unwrap_or_default();
+                line.push_str(&format!(",{}", escape_csv_field(&utm_src)));
+                col_idx += 1;
+            }
+            if has_utm_campaign {
+                let utm_camp: String = row.get(col_idx).unwrap_or_default();
+                line.push_str(&format!(",{}", escape_csv_field(&utm_camp)));
+            }
+            line.push('\n');
+
+            csv_buffer.push_str(&line);
+            if csv_buffer.len() >= 8192 {
+                let bytes = axum::body::Bytes::from(csv_buffer);
+                if tx.blocking_send(Ok(bytes)).is_err() {
+                    return;
+                }
+                csv_buffer = String::new();
+            }
+        }
+
+        if !csv_buffer.is_empty() {
+            let _ = tx.blocking_send(Ok(axum::body::Bytes::from(csv_buffer)));
+        }
+    });
+
+    let stream = DbExportStream { receiver: rx };
+    let filename = if target_type == "url" {
+        format!("url_{}_analytics.csv", id)
+    } else {
+        format!("page_{}_analytics.csv", id)
+    };
+
+    (
+        StatusCode::OK,
+        [
+            ("Content-Type", "text/csv"),
+            (
+                "Content-Disposition",
+                &format!("attachment; filename=\"{}\"", filename),
+            ),
+            ("X-BZOD-Export-Records", &count.to_string()),
+        ],
+        axum::body::Body::from_stream(stream),
+    )
+        .into_response()
+}
+pub(crate) async fn perform_json_export(
+    state: AppState,
+    target_type: &'static str,
+    id: String,
+    date_from: Option<String>,
+    date_to: Option<String>,
+) -> Response {
+    let (clean_date_from, clean_date_to) =
+        match validate_date_filters(date_from.as_deref(), date_to.as_deref()) {
+            Ok(res) => res,
+            Err(status) => return status.into_response(),
+        };
+
+    let target_exists = {
+        let conn = state.content_db.lock().unwrap();
+        if target_type == "url" {
+            get_url_by_id(&conn, &id)
+                .map(|u| u.is_some())
+                .unwrap_or(false)
+        } else {
+            get_landing_page_by_id(&conn, &id)
+                .map(|p| p.is_some())
+                .unwrap_or(false)
+        }
+    };
+    if !target_exists {
+        return (StatusCode::NOT_FOUND, "Target not found").into_response();
+    }
+
+    let count = {
+        let conn = state.analytics_db.lock().unwrap();
+        get_target_visit_total_filtered(
+            &conn,
+            target_type,
+            &id,
+            clean_date_from.as_deref(),
+            clean_date_to.as_deref(),
+        )
+        .unwrap_or(0)
+    };
+
+    if count > MAX_JSON_EXPORT_ROWS as i64 {
+        return StatusCode::PAYLOAD_TOO_LARGE.into_response();
+    }
+
+    let visits_raw = {
+        let conn = state.analytics_db.lock().unwrap();
+        match get_target_visits_all_in_memory(
+            &conn,
+            target_type,
+            &id,
+            clean_date_from.as_deref(),
+            clean_date_to.as_deref(),
+        ) {
+            Ok(v) => v,
+            Err(_) => return (StatusCode::INTERNAL_SERVER_ERROR, "Database error").into_response(),
+        }
+    };
+
+    #[derive(serde::Serialize)]
+    struct JsonExportRow {
+        timestamp: String,
+        ip_address: String,
+        country: String,
+        referrer: String,
+        browser: String,
+        user_agent: String,
+    }
+
+    let export_rows: Vec<JsonExportRow> = visits_raw
+        .into_iter()
+        .map(|r| {
+            let (browser, _, _) = parse_ua(&r.user_agent);
+            let referrer = clean_referrer(&r.referer);
+            let country_display = if r.country.is_empty() {
+                "Unknown".to_string()
+            } else {
+                r.country
+            };
+            JsonExportRow {
+                timestamp: r.timestamp,
+                ip_address: r.ip_address,
+                country: country_display,
+                referrer,
+                browser,
+                user_agent: r.user_agent,
+            }
+        })
+        .collect();
+
+    let body_str = match serde_json::to_string(&export_rows) {
+        Ok(s) => s,
+        Err(_) => {
+            return (StatusCode::INTERNAL_SERVER_ERROR, "Serialization error").into_response()
+        }
+    };
+
+    let filename = if target_type == "url" {
+        format!("url_{}_analytics.json", id)
+    } else {
+        format!("page_{}_analytics.json", id)
+    };
+
+    (
+        StatusCode::OK,
+        [
+            ("Content-Type", "application/json"),
+            (
+                "Content-Disposition",
+                &format!("attachment; filename=\"{}\"", filename),
+            ),
+            ("X-BZOD-Export-Records", &count.to_string()),
+        ],
+        body_str,
+    )
+        .into_response()
+}
+// Helpers
+pub(crate) fn format_size(bytes: u64) -> String {
+    if bytes < 1024 {
+        format!("{} B", bytes)
+    } else if bytes < 1024 * 1024 {
+        format!("{:.2} KB", bytes as f64 / 1024.0)
+    } else {
+        format!("{:.2} MB", bytes as f64 / (1024.0 * 1024.0))
+    }
+}
+pub(crate) fn get_dir_size(dir: &std::path::Path) -> std::io::Result<u64> {
+    let mut total = 0;
+    if dir.is_dir() {
+        for entry in std::fs::read_dir(dir)? {
+            let entry = entry?;
+            let path = entry.path();
+            if path.is_dir() {
+                total += get_dir_size(&path)?;
+            } else {
+                total += entry.metadata()?.len();
+            }
+        }
+    }
+    Ok(total)
+}
+pub(crate) async fn perform_user_csv_export(
+    user_dbs: crate::state::UserDbs,
+    target_type: &'static str,
+    id: String,
+    date_from: Option<String>,
+    date_to: Option<String>,
+) -> Response {
+    let (clean_date_from, clean_date_to) =
+        match validate_date_filters(date_from.as_deref(), date_to.as_deref()) {
+            Ok(res) => res,
+            Err(status) => return status.into_response(),
+        };
+
+    let target_exists = {
+        let conn = user_dbs.content.lock().unwrap();
+        if target_type == "url" {
+            get_url_by_id(&conn, &id)
+                .map(|u| u.is_some())
+                .unwrap_or(false)
+        } else {
+            get_landing_page_by_id(&conn, &id)
+                .map(|p| p.is_some())
+                .unwrap_or(false)
+        }
+    };
+    if !target_exists {
+        return (StatusCode::NOT_FOUND, "Target not found").into_response();
+    }
+
+    let count = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        get_target_visit_total_filtered(
+            &conn,
+            target_type,
+            &id,
+            clean_date_from.as_deref(),
+            clean_date_to.as_deref(),
+        )
+        .unwrap_or(0)
+    };
+
+    let (has_utm_source, has_utm_campaign) = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        let cols = get_visits_schema_columns(&conn).unwrap_or_default();
+        (cols.contains("utm_source"), cols.contains("utm_campaign"))
+    };
+
+    let (tx, rx) =
+        tokio::sync::mpsc::channel::<Result<axum::body::Bytes, std::convert::Infallible>>(32);
+    let analytics_db = user_dbs.analytics.clone();
+    let target_id = id.clone();
+
+    tokio::task::spawn_blocking(move || {
+        let conn = analytics_db.lock().unwrap();
+
+        let mut header = "Timestamp,IP Address,Country,Referrer,Browser,User-Agent".to_string();
+        if has_utm_source {
+            header.push_str(",UTM Source");
+        }
+        if has_utm_campaign {
+            header.push_str(",UTM Campaign");
+        }
+        header.push('\n');
+
+        if tx
+            .blocking_send(Ok(axum::body::Bytes::from(header)))
+            .is_err()
+        {
+            return;
+        }
+
+        let select_fields = if has_utm_source && has_utm_campaign {
+            "timestamp, ip_address, country, referer, user_agent, utm_source, utm_campaign"
+        } else if has_utm_source {
+            "timestamp, ip_address, country, referer, user_agent, utm_source"
+        } else if has_utm_campaign {
+            "timestamp, ip_address, country, referer, user_agent, utm_campaign"
+        } else {
+            "timestamp, ip_address, country, referer, user_agent"
+        };
+
+        let mut sql = format!(
+            "SELECT {} FROM visits WHERE target_type = ?1 AND target_id = ?2",
+            select_fields
+        );
+        let mut params: Vec<Box<dyn rusqlite::ToSql>> =
+            vec![Box::new(target_type.to_string()), Box::new(target_id)];
+
+        if let Some(df) = clean_date_from.as_deref() {
+            sql.push_str(&format!(" AND timestamp >= ?{}", params.len() + 1));
+            params.push(Box::new(format!("{}T00:00:00Z", df)));
+        }
+
+        if let Some(dt) = clean_date_to.as_deref() {
+            if let Ok(parsed_date) = chrono::NaiveDate::parse_from_str(dt, "%Y-%m-%d") {
+                let next_day = parsed_date + chrono::Duration::days(1);
+                sql.push_str(&format!(" AND timestamp < ?{}", params.len() + 1));
+                params.push(Box::new(format!(
+                    "{}T00:00:00Z",
+                    next_day.format("%Y-%m-%d")
+                )));
+            }
+        }
+
+        sql.push_str(" ORDER BY timestamp DESC, id DESC");
+
+        let mut stmt = match conn.prepare(&sql) {
+            Ok(s) => s,
+            Err(_) => return,
+        };
+
+        let param_refs: Vec<&dyn rusqlite::ToSql> = params.iter().map(|p| p.as_ref()).collect();
+        let mut rows = match stmt.query(rusqlite::params_from_iter(param_refs)) {
+            Ok(r) => r,
+            Err(_) => return,
+        };
+
+        let mut csv_buffer = String::new();
+
+        while let Ok(Some(row)) = rows.next() {
+            let timestamp: String = row.get(0).unwrap_or_default();
+            let ip_address: String = row.get(1).unwrap_or_default();
+            let country: String = row.get(2).unwrap_or_default();
+            let referer: String = row.get(3).unwrap_or_default();
+            let user_agent: String = row.get(4).unwrap_or_default();
+
+            let (browser, _, _) = parse_ua(&user_agent);
+            let referrer = clean_referrer(&referer);
+            let country_display = if country.is_empty() {
+                "Unknown".to_string()
+            } else {
+                country
+            };
+
+            let mut line = format!(
+                "{},{},{},{},{},{}",
+                escape_csv_field(&timestamp),
+                escape_csv_field(&ip_address),
+                escape_csv_field(&country_display),
+                escape_csv_field(&referrer),
+                escape_csv_field(&browser),
+                escape_csv_field(&user_agent)
+            );
+
+            let mut col_idx = 5;
+            if has_utm_source {
+                let utm_src: String = row.get(col_idx).unwrap_or_default();
+                line.push_str(&format!(",{}", escape_csv_field(&utm_src)));
+                col_idx += 1;
+            }
+            if has_utm_campaign {
+                let utm_camp: String = row.get(col_idx).unwrap_or_default();
+                line.push_str(&format!(",{}", escape_csv_field(&utm_camp)));
+            }
+            line.push('\n');
+
+            csv_buffer.push_str(&line);
+            if csv_buffer.len() >= 8192 {
+                let bytes = axum::body::Bytes::from(csv_buffer);
+                if tx.blocking_send(Ok(bytes)).is_err() {
+                    return;
+                }
+                csv_buffer = String::new();
+            }
+        }
+
+        if !csv_buffer.is_empty() {
+            let _ = tx.blocking_send(Ok(axum::body::Bytes::from(csv_buffer)));
+        }
+    });
+
+    let stream = DbExportStream { receiver: rx };
+    let filename = if target_type == "url" {
+        format!("url_{}_analytics.csv", id)
+    } else {
+        format!("page_{}_analytics.csv", id)
+    };
+
+    (
+        StatusCode::OK,
+        [
+            ("Content-Type", "text/csv"),
+            (
+                "Content-Disposition",
+                &format!("attachment; filename=\"{}\"", filename),
+            ),
+            ("X-BZOD-Export-Records", &count.to_string()),
+        ],
+        axum::body::Body::from_stream(stream),
+    )
+        .into_response()
+}
+pub(crate) async fn perform_user_json_export(
+    user_dbs: crate::state::UserDbs,
+    target_type: &'static str,
+    id: String,
+    date_from: Option<String>,
+    date_to: Option<String>,
+) -> Response {
+    let (clean_date_from, clean_date_to) =
+        match validate_date_filters(date_from.as_deref(), date_to.as_deref()) {
+            Ok(res) => res,
+            Err(status) => return status.into_response(),
+        };
+
+    let target_exists = {
+        let conn = user_dbs.content.lock().unwrap();
+        if target_type == "url" {
+            get_url_by_id(&conn, &id)
+                .map(|u| u.is_some())
+                .unwrap_or(false)
+        } else {
+            get_landing_page_by_id(&conn, &id)
+                .map(|p| p.is_some())
+                .unwrap_or(false)
+        }
+    };
+    if !target_exists {
+        return (StatusCode::NOT_FOUND, "Target not found").into_response();
+    }
+
+    let count = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        get_target_visit_total_filtered(
+            &conn,
+            target_type,
+            &id,
+            clean_date_from.as_deref(),
+            clean_date_to.as_deref(),
+        )
+        .unwrap_or(0)
+    };
+
+    if count > MAX_JSON_EXPORT_ROWS as i64 {
+        return StatusCode::PAYLOAD_TOO_LARGE.into_response();
+    }
+
+    let visits_raw = {
+        let conn = user_dbs.analytics.lock().unwrap();
+        match get_target_visits_all_in_memory(
+            &conn,
+            target_type,
+            &id,
+            clean_date_from.as_deref(),
+            clean_date_to.as_deref(),
+        ) {
+            Ok(v) => v,
+            Err(_) => return (StatusCode::INTERNAL_SERVER_ERROR, "Database error").into_response(),
+        }
+    };
+
+    #[derive(serde::Serialize)]
+    struct JsonExportRow {
+        timestamp: String,
+        ip_address: String,
+        country: String,
+        referrer: String,
+        browser: String,
+        user_agent: String,
+    }
+
+    let export_rows: Vec<JsonExportRow> = visits_raw
+        .into_iter()
+        .map(|r| {
+            let (browser, _, _) = parse_ua(&r.user_agent);
+            let referrer = clean_referrer(&r.referer);
+            let country_display = if r.country.is_empty() {
+                "Unknown".to_string()
+            } else {
+                r.country
+            };
+            JsonExportRow {
+                timestamp: r.timestamp,
+                ip_address: r.ip_address,
+                country: country_display,
+                referrer,
+                browser,
+                user_agent: r.user_agent,
+            }
+        })
+        .collect();
+
+    let body_str = match serde_json::to_string(&export_rows) {
+        Ok(s) => s,
+        Err(_) => {
+            return (StatusCode::INTERNAL_SERVER_ERROR, "Serialization error").into_response()
+        }
+    };
+
+    let filename = if target_type == "url" {
+        format!("url_{}_analytics.json", id)
+    } else {
+        format!("page_{}_analytics.json", id)
+    };
+
+    (
+        StatusCode::OK,
+        [
+            ("Content-Type", "application/json"),
+            (
+                "Content-Disposition",
+                &format!("attachment; filename=\"{}\"", filename),
+            ),
+            ("X-BZOD-Export-Records", &count.to_string()),
+        ],
+        body_str,
+    )
+        .into_response()
+}
+
+// --- Feature modules ---
+
+mod auth;
+pub use auth::*;
+mod dashboard;
+pub use dashboard::*;
+mod urls;
+pub use urls::*;
+mod pages;
+pub use pages::*;
+mod users;
+pub use users::*;
+mod analytics;
+pub use analytics::*;
+mod settings;
+pub use settings::*;
+mod audit;
+pub use audit::*;
+mod sessions;
+pub use sessions::*;
+mod quotas;
+pub use quotas::*;
+mod health;
+pub use health::*;
+mod backups;
+pub use backups::*;
+mod api_keys;
+pub use api_keys::*;
+mod moderation;
+pub use moderation::*;
Author	SHA1	Message	Date
thakares	f49698bb5c	Release v0.6.0 Rust CI / Build Docker Image (push) Canceled after 0s Details Rust CI / Test & Quality Checks (push) Canceled after 0s Details	2026-08-09 17:17:57 +05:30
thakares	7069ca9db7	docs(web): update social preview image	2026-07-01 15:15:08 +05:30
thakares	faf8fc0eda	docs(web): update social preview image	2026-07-01 15:12:47 +05:30
thakares	bf29ccab56	fix(web): correct Open Graph preview image URL	2026-07-01 14:51:27 +05:30
thakares	667503c8f6	fix(web): add social preview image assets	2026-07-01 14:42:10 +05:30
thakares	83218ba602	docs: update documentation for v0.5.3 - Refresh administrator guide - Update architecture documentation - Document Registry Validator - Document Registry Repair Framework - Update backup and restore guide - Refresh CLI documentation - Update Docker deployment example - Bump version to v0.5.3	2026-06-29 17:44:10 +05:30
thakares	58c0af6510	docs: refresh README for v0.5.3 - Rewrite project overview - Update architecture documentation - Document Registry Validator and Repair Framework - Refresh CLI reference - Expand installation and deployment guides - Update feature matrix - Add roadmap and operational tooling - Update badges and version	2026-06-29 17:12:43 +05:30
thakares	f4947489af	feat: add registry repair framework - add shared registry validation service - add transaction-safe repair CLI - refactor doctor to use validator - improve restore integrity checks - add registry repair integration tests - strengthen global slug consistency	2026-06-29 16:53:05 +05:30
thakares	2761863c14	Release v0.5.2 - Add admin content consistency diagnostics - Add admin-migrate CLI - Harden RBAC for API endpoints - Normalize multi-tenant storage paths - Improve backup and restore compatibility - Fix administrator routing consistency - Improve doctor and stats commands	2026-06-29 16:11:43 +05:30
thakares	a32c0fd7ca	docs: update README and v0.5.1 release notes	2026-06-20 20:58:49 +05:30
thakares	b03e0ea727	Release v0.5.1: namespace integrity and multi-user hardening	2026-06-20 20:18:11 +05:30
thakares	115f6e9a23	Release v0.5.1: namespace integrity, dashboard parity and QR hardening	2026-06-20 19:54:29 +05:30
thakares	0295b4bd7c	www/index.html page updated as per the refactoring...	2026-06-19 22:00:18 +05:30
thakares	6a3c744667	www/index.html page updated as per the refactoring...	2026-06-19 20:58:53 +05:30
thakares	19e48270ed	docs: update landing page UI and add comparison tables	2026-06-19 19:59:58 +05:30
thakares	133c707f27	docs: update landing page UI and add comparison tables	2026-06-19 19:36:22 +05:30
thakares	496186e2af	docs: complete v0.5.0 administration, architecture and deployment guides	2026-06-19 16:19:31 +05:30
thakares	fe7e8efeef	docs: complete v0.5.0 administration, architecture and deployment guides	2026-06-19 16:10:48 +05:30
thakares	5193870c97	docs: complete v0.5.0 administration, architecture and deployment guides	2026-06-19 16:07:11 +05:30
thakares	7fdc352547	docs: complete v0.5.0 administration, architecture and deployment guides	2026-06-19 16:05:43 +05:30
thakares	49acf76cf6	docs: complete v0.5.0 administration, architecture and deployment guides	2026-06-19 15:58:23 +05:30
thakares	c7e851000f	docs: Release Notes on BZOD v0.5.0 — General Availability	2026-06-19 15:16:13 +05:30
thakares	c5f33e9713	docs: expand README and testing documentation	2026-06-19 15:07:43 +05:30
thakares	7dfb8c0f1b	BZOD v0.5.0 RC2: multi-user platform, dashboards, analytics, backups and validation	2026-06-19 14:51:38 +05:30
thakares	743502b183	ci: improve Rust workflow checks	2026-06-18 15:31:52 +05:30
thakares	7ee6ab2feb	ci: rerun workflow	2026-06-18 15:25:06 +05:30
thakares	621a1eccdc	Add project comparison guide and improve landing page	2026-06-18 15:09:09 +05:30
thakares	cefb84f643	Add project comparison guide and improve landing page	2026-06-18 15:02:31 +05:30
thakares	9fae39dfa4	Use standard Rust dual-license layout	2026-06-18 13:07:34 +05:30
thakares	2212701b6b	Add Apache 2.0 license file	2026-06-18 13:04:28 +05:30
thakares	536d885a3d	Add API documentation and dual MIT/Apache licensing	2026-06-18 12:59:07 +05:30
thakares	a4ffe9a509	Delete LICENSE-APACHE	2026-06-18 12:39:03 +05:30
thakares	ad05af95e9	Add API documentation and dual MIT/Apache licensing	2026-06-18 12:32:51 +05:30
thakares	17d8618443	docs: add REST API documentation and installation guide	2026-06-18 12:17:16 +05:30
thakares	74524cb7d2	docs: add one-command deployment section with deploy.sh	2026-06-17 20:45:54 +05:30
thakares	e2140e6614	Add deploy script endpoint and improve landing page	2026-06-17 20:33:25 +05:30
thakares	1a9bf096f3	Add deploy.sh endpoint and one-command installation flow	2026-06-17 20:17:19 +05:30
thakares	900f72a299	release: update installer	2026-06-17 19:08:19 +05:30
thakares	d42f6da94a	feat: production deployment script with rollback and multi-arch support	2026-06-17 19:00:41 +05:30
thakares	18d8b16a2c	docs: update documentation for v0.4.0 release	2026-06-17 16:49:47 +05:30
thakares	f6dcf58b79	Add analytics drill-down, visitor logs, exports and pagination	2026-06-17 15:47:14 +05:30
thakares	84c48fa5c1	Add analytics drill-down, visitor logs, exports and pagination	2026-06-17 15:35:59 +05:30
thakares	0e111d61ff	docs: add Docker deployment guide	2026-06-14 21:02:36 +05:30
thakares	81eb8950dd	ci: add automatic Docker image push to GHCR on main	2026-06-14 20:53:32 +05:30
thakares	914563e883	ci: fix Docker image loading in CI health check	2026-06-14 20:49:59 +05:30
thakares	8b521f1a71	ci: fix YAML structure in GitHub workflow	2026-06-14 20:46:55 +05:30
thakares	f1d72c8cbe	Update rust.yml	2026-06-14 20:40:22 +05:30
thakares	d9979ba04c	docs: refine README and project documentation	2026-06-14 19:50:12 +05:30
thakares	6f42b43608	docs: refine README and project documentation	2026-06-14 19:44:36 +05:30
thakares	3e9dc47604	docs: refine README and project documentation	2026-06-14 19:43:46 +05:30
thakares	e2e61fc412	docs: refine README and project documentation	2026-06-14 19:42:35 +05:30
thakares	a0a73b918c	docs: refine README and project documentation	2026-06-14 19:36:29 +05:30
thakares	d2174aa111	docs: refine README and project documentation	2026-06-14 19:26:21 +05:30
thakares	6a45474e97	docs: refine README and project documentation	2026-06-14 19:24:38 +05:30
thakares	c6486763e3	Add custom slugs, restore UI, UTM builder, and CLI link tools	2026-06-14 19:11:21 +05:30
thakares	02a26cfd94	style: fix cargo fmt issues in pages.rs and root_landing_tests.rs	2026-06-13 22:29:03 +05:30
thakares	3c0a1bdd91	ci: enhance GitHub workflow with Docker build + better Rust CI	2026-06-13 22:24:37 +05:30
thakares	ee6d3135d5	docker: optimize build cache with better layer ordering	2026-06-13 22:20:25 +05:30
thakares	671370571a	chore: add .dockerignore for cleaner Docker builds	2026-06-13 22:13:28 +05:30
thakares	42a2df33dd	Add root landing page support and package static assets	2026-06-13 21:47:14 +05:30
thakares	80038fb851	docs: improve testing guide and add README documentation links	2026-06-13 18:48:26 +05:30
thakares	9c5e3e4d58	Add TESTING.md with validation and recovery procedures	2026-06-13 18:35:05 +05:30