With ~1000 posts going back to 2003, a lot of them are dated in a way that either doesn't matter (personal essays that age fine) or actively hurts (event announcements, reactions to now-dead external links). This is the frontmatter/tooling used to triage that without deleting anything.
| Field | Effect | Where it's used |
|---|---|---|
noindex: true |
Adds <meta name="robots" content="noindex">. Post stays fully visible in /blog/, RSS, tag pages, related-posts, and on-site search — it's just excluded from search engines and from sitemap.xml. |
layouts/partials/extra-head.html, layouts/sitemap.xml |
historical: true |
Renders a small "kept for historical/nostalgic interest" banner under the title. No build or SEO effect. | layouts/posts/single.html |
draft: true + file moved to archived-posts/ |
Fully excluded from npm run build (Hugo's buildDrafts = false) and physically out of content/Blog/, so day-to-day the folder only has live/pending posts. |
Hugo core (config.toml), archived-posts/ (outside contentDir) |
curated: <date> |
No rendering effect — just a marker that a post has already been reviewed by this project, so a future pass can tell what's already been triaged. | none (bookkeeping only) |
hidden: true (pre-existing, unrelated to this project) |
noindex and removed from /blog/ listing, RSS, and related-posts. Used on 3 personal posts. Stronger than noindex: true above — don't confuse the two. |
themes/hello-friend-ng/layouts/partials/head.html, _default/list.html, _default/rss.xml, partials/related.html |
node scripts/curate-posts.js noindex <slug...>
node scripts/curate-posts.js un-noindex <slug...>
node scripts/curate-posts.js historical <slug...>
node scripts/curate-posts.js un-historical <slug...>
node scripts/curate-posts.js archive <slug...>
node scripts/curate-posts.js restore <slug...>
node scripts/curate-posts.js reviewed <slug...>reviewed just stamps curated: <today> with no other change — for the "leave alone / evergreen" bucket, which doesn't get a field of its own otherwise.
A slug is the filename without .md, e.g. spiderman-2-the-game-not-the-movie. Every command also accepts --file <path> for batch mode — a text file with one slug per line (blank lines and # comments ignored), which is how the bulk triage report gets applied.
archive sets draft: true + archivedDate and git mvs the file into archived-posts/. restore reverses that exactly — moves it back to content/Blog/ and drops both fields. Both preserve git history via git mv (falls back to a plain filesystem move, with a warning to git add manually, if the file isn't tracked).
After archiving posts, run npm run og-images:prune to delete their now-orphaned social-card images (it already skips draft: true posts and prunes anything not referenced by the current file set — no special-casing needed for archived-posts/, it's just outside content/Blog so it's already invisible to that script).
Changed your mind about an archived post? node scripts/curate-posts.js restore <slug> and it's back exactly where it was. Same idea for un-noindex/un-historical.