refactor: split site and api apps

.github/workflows/ci.yml

@@ -36,10 +36,10 @@ jobs:
       - name: Build
         run: bun run build
 
-      - name: Type check (web)
-        run: bun run --cwd apps/web typecheck
+      - name: Type check (site)
+        run: bun run --cwd apps/site typecheck
 
-      - name: Build (web)
+      - name: Build (site)
         env:
           DATABASE_URL: ${{ secrets.DATABASE_URL }}
-        run: bun run --cwd apps/web build
+        run: bun run --cwd apps/site build

.github/workflows/deploy-site.yml

@@ -6,8 +6,7 @@ on:
     branches: [main]
     paths:
       - ".github/workflows/deploy-site.yml"
-      - "apps/web/**"
-      - "!apps/web/worker/**"
+      - "apps/site/**"
 
 jobs:
   deploy:
@@ -28,7 +27,7 @@ jobs:
       - name: Build site
         env:
           DATABASE_URL: ${{ secrets.DATABASE_URL }}
-        run: bun run --cwd apps/web build
+        run: bun run --cwd apps/site build
 
       - name: Upload site to R2
         env:
@@ -45,15 +44,15 @@ jobs:
           # with stale-while-revalidate for 1 hour. A Cloudflare Cache Rule on
           # the zone enables caching for these specific paths (extensionless
           # HTML isn't cached by Cloudflare default); see infrastructure notes
-          # in apps/web/README.md. Static assets (favicons, og-image, logo)
+          # in apps/site/README.md. Static assets (favicons, og-image, logo)
           # keep Cloudflare's default static-asset cache behavior - no header.
           CACHE_HTML="public, max-age=300, stale-while-revalidate=3600"
 
           # Non-HTML assets: copy the finite Astro output directly. Do not use
           # `aws s3 sync` at the bucket root because the same bucket also stores
           # the million-plus scraper-managed /documents/ and /extracted/ objects.
-          find apps/web/dist -type f ! -name "*.html" -print0 | while IFS= read -r -d '' f; do
-            rel="${f#apps/web/dist/}"
+          find apps/site/dist -type f ! -name "*.html" -print0 | while IFS= read -r -d '' f; do
+            rel="${f#apps/site/dist/}"
             case "$rel" in
               sitemap.xml|robots.txt|llms.txt)
                 aws s3 cp "$f" "$R2_BUCKET/$rel" \
@@ -70,8 +69,8 @@ jobs:
           # HTML files: upload to extensionless keys to match canonical URLs.
           # /classification.html -> r2://classification, /types/legal.html -> r2://types/legal.
           # index.html is the one exception; it stays as a static homepage at /.
-          find apps/web/dist -name "*.html" -print0 | while IFS= read -r -d '' f; do
-            rel="${f#apps/web/dist/}"
+          find apps/site/dist -name "*.html" -print0 | while IFS= read -r -d '' f; do
+            rel="${f#apps/site/dist/}"
             if [ "$rel" = "index.html" ]; then
               key="index.html"
             else

.github/workflows/deploy-worker.yml

@@ -5,7 +5,7 @@ on:
   push:
     branches: [main]
     paths:
-      - "apps/web/worker/**"
+      - "apps/api/**"
 
 jobs:
   deploy:
@@ -16,14 +16,14 @@ jobs:
 
       - uses: oven-sh/setup-bun@v2
 
-      - run: bun install --cwd apps/web/worker
+      - run: bun install --cwd apps/api
 
       - name: Deploy to Cloudflare Workers
         uses: cloudflare/wrangler-action@v3
         with:
           apiToken: ${{ secrets.CLOUDFLARE_API_TOKEN }}
           accountId: ${{ secrets.CLOUDFLARE_ACCOUNT_ID }}
-          workingDirectory: apps/web/worker
+          workingDirectory: apps/api
           secrets: |
             DATABASE_URL
         env:

.gitignore

@@ -17,11 +17,11 @@ dist/
 corpus/
 dev/
 
-# Generated web data (local only)
-apps/web/data/
+# Generated site data (local only)
+apps/site/data/
 
 # Astro
-apps/web/.astro/
+apps/site/.astro/
 
 # Tests
 coverage/

README.md

@@ -66,7 +66,8 @@ Run `corpus <command> --help` for detailed options.
 apps/
   cli/              # Unified CLI — corpus <command>
   cdx-filter/       # AWS Lambda — filters CDX indexes for .docx URLs
-  web/              # Landing page (docxcorp.us) + Cloudflare Worker API
+  site/             # Static Astro site for docxcorp.us
+  api/              # Cloudflare Worker API for api.docxcorp.us
 packages/
   shared/           # DB client, storage abstraction, formatting
   scraper/          # Downloads WARC, validates .docx, deduplicates
@@ -84,7 +85,8 @@ db/
 |-------|------|---------|
 | **cli** | `corpus` command — orchestrates everything | Bun |
 | **cdx-filter** | Filter Common Crawl CDX indexes (Lambda) | Node.js |
-| **web** | docxcorp.us landing page + API worker | Static + CF Worker |
+| **site** | docxcorp.us landing page and dataset pages | Static Astro |
+| **api** | api.docxcorp.us `/stats`, `/documents`, `/manifest` | Cloudflare Worker |
 | **scraper** | Download, validate, deduplicate .docx files | Bun |
 | **extractor** | Extract text + detect language (Docling) | Bun + Python |
 | **embedder** | Generate embeddings (Gemini) | Bun |
@@ -244,8 +246,8 @@ docker compose up -d
 DATABASE_URL=postgres://postgres:postgres@localhost:5432/docx_corpus \
   bun run corpus status
 
-# Run web API locally
-cd apps/web/worker
+# Run API locally
+cd apps/api
 npx wrangler dev
 ```
 

apps/web/README.md → apps/site/README.md

@@ -1,4 +1,4 @@
-# docxcorp.us web
+# docxcorp.us site
 
 Static site assets are deployed to Cloudflare R2 by `.github/workflows/deploy-site.yml`.
 

apps/web/package.json → apps/site/package.json

@@ -1,5 +1,5 @@
 {
-  "name": "@docx-corpus/web",
+  "name": "@docx-corpus/site",
   "private": true,
   "version": "0.0.0",
   "type": "module",

apps/web/src/lib/content.ts → apps/site/src/lib/content.ts

@@ -1,5 +1,5 @@
 /**
- * Content loader for markdown drafts in apps/web/content/.
+ * Content loader for markdown drafts in apps/site/content/.
  *
  * Each .md file has YAML frontmatter parsed by Astro's import.meta.glob.
  * The body is rendered to HTML at build time.

apps/web/src/pages/index.astro → apps/site/src/pages/index.astro

@@ -1,6 +1,6 @@
 ---
-// Homepage. Ported from the legacy static apps/web/public/index.html so it
-// participates in the Astro build pipeline. Inline CSS and JS are preserved
+// Homepage. Ported from the legacy static file so it participates in the
+// Astro build pipeline. Inline CSS and JS are preserved
 // verbatim under `is:global is:inline` and `is:inline` respectively; the
 // explorer JS still fetches live data from api.docxcorp.us at runtime, same
 // as before.
@@ -1564,4 +1564,4 @@ curl "https://api.docxcorp.us/manifest?type=legal&lang=en&min_confidence
 
 </body>
 
-</html>
+</html>

package.json

@@ -2,7 +2,9 @@
   "name": "docx-corpus",
   "private": true,
   "workspaces": [
-    "apps/*",
+    "apps/cdx-filter",
+    "apps/cli",
+    "apps/site",
     "packages/*"
   ],
   "scripts": {
@@ -16,11 +18,13 @@
     "release:cli": "bun run --cwd apps/cli release",
     "setup:extractor": "bun run --cwd packages/extractor setup",
     "prepare": "lefthook install",
-    "dev:web": "bun run --cwd apps/web dev",
-    "build:web": "bun run --cwd apps/web build"
+    "dev:site": "bun run --cwd apps/site dev",
+    "build:site": "bun run --cwd apps/site build",
+    "dev:web": "bun run dev:site",
+    "build:web": "bun run build:site"
   },
   "devDependencies": {
     "@biomejs/biome": "^2.4.6",
     "lefthook": "^1.11.13"
   }
-}
+}