Refactor PDF generation script

Author: webstackdev

.clinerules

@@ -27,6 +27,16 @@ There is no stop button to break out of a hung loop.
   possible, pass `--yes` / `-y` (e.g. `npx --yes <pkg>`).
 - **`gh`**: use `--no-pager` where supported and avoid interactive subcommands.
 
+### Git is read-only for the agent
+
+NEVER run state-changing git commands: `git commit`, `git add`, `git push`,
+`git stash`, `git checkout`, `git switch`, `git restore`, `git reset`,
+`git rebase`, `git merge`, `git cherry-pick`, `git tag`, etc. The user handles
+all git mutations themselves — do not commit work, even when a task mirrors a
+previously committed step. Read-only inspection is fine, always with
+`--no-pager` per the rules above (`git status`, `git --no-pager log`,
+`git --no-pager diff`, `git --no-pager show`, `git --no-pager blame`).
+
 ### General rules for `execute_command`
 
 - Every command must be **non-interactive** and **self-terminating**.

.github/workflows/search.yml

@@ -66,7 +66,7 @@ jobs:
         run: |
           COLLECTIONS=""
           if [ "${{ steps.scope.outputs.crawl_articles }}" = "true" ]; then
-            COLLECTIONS="$COLLECTIONS --collection articles --collection deep-dive"
+            COLLECTIONS="$COLLECTIONS --collection articles"
           fi
           if [ "${{ steps.scope.outputs.crawl_services }}" = "true" ]; then
             COLLECTIONS="$COLLECTIONS --collection services"

scripts/generate-pdfs/index.mjs

@@ -4,8 +4,8 @@
  * Generate PDFs from /print/ pages using Puppeteer.
  *
  * Usage:
- * - node scripts/generate-pdfs.mjs <slug>       # single article
- * - node scripts/generate-pdfs.mjs              # all deep dive articles
+ * - node scripts/generate-pdfs/index.mjs <slug>       # single article
+ * - node scripts/generate-pdfs/index.mjs              # all articles
  * - OR npm run pdf:generate
  *
  * Requires a running server (dev or preview) at localhost:4321.
@@ -60,7 +60,7 @@ const LAYOUT_SETTLE_DELAY_MS = 500
 const collectAllSlugs = () => {
   if (!existsSync(ARTICLES_DIR)) return []
   return readdirSync(ARTICLES_DIR, { withFileTypes: true })
-    .filter(e => e.isDirectory() && existsSync(join(ARTICLES_DIR, e.name, 'pdf.mdx')))
+    .filter(e => e.isDirectory() && existsSync(join(ARTICLES_DIR, e.name, 'index.mdx')))
     .map(e => e.name)
     .sort()
 }
@@ -279,18 +279,18 @@ if (!(await checkServer())) {
 }
 
 if (slugFilter) {
-  const pdfMdx = join(ARTICLES_DIR, slugFilter, 'pdf.mdx')
-  if (!existsSync(pdfMdx)) {
-    console.error(`ERROR: No pdf.mdx found for "${slugFilter}"`)
-    console.error(`Expected: ${pdfMdx}`)
+  const indexMdx = join(ARTICLES_DIR, slugFilter, 'index.mdx')
+  if (!existsSync(indexMdx)) {
+    console.error(`ERROR: No index.mdx found for "${slugFilter}"`)
+    console.error(`Expected: ${indexMdx}`)
     process.exit(1)
   }
 }
 
 const slugs = slugFilter ? [slugFilter] : collectAllSlugs()
 
 if (slugs.length === 0) {
-  console.error('ERROR: No deep dive articles with pdf.mdx found.')
+  console.error('ERROR: No articles with index.mdx found.')
   process.exit(1)
 }
 

scripts/search-index.py

@@ -45,7 +45,6 @@ class CollectionConfig:
 # Collections to index
 COLLECTIONS: Final[list[CollectionConfig]] = [
   CollectionConfig(name="articles", url_prefix="/articles", source_dir="articles", glob_pattern="**/index.mdx"),
-  CollectionConfig(name="deep-dive", url_prefix="/deep-dive", source_dir="articles", glob_pattern="**/pdf.mdx"),
   CollectionConfig(name="case-studies", url_prefix="/case-studies", source_dir="case-studies", glob_pattern="**/index.mdx"),
   CollectionConfig(name="services", url_prefix="/services", source_dir="services", glob_pattern="**/index.md"),
 ]