Write SEO blog posts for shipped-but-unblogged features

[vivek7405]

Problem

The blog/ directory has 29 posts but a batch of shipped, SEO-worthy features have no post yet. These are high search-intent topics (Next.js-alternative migration queries, concrete how-tos) that should drive organic traffic to the marketing site.

Design / approach

Write one markdown post per topic in blog/<slug>.md, matching the existing house style (first-person, plain-language hook, jargon explained in parentheses, no em-dashes per AGENTS.md invariant 11, SEO-rich description front matter). Slot dates into the existing timeline (latest existing post is 2026-07-06) and lightly adjust a few existing dates so the cadence reads evenly. All posts land in one PR.

Topics (each backed by a shipped PR):

• Next.js 16 file-routing parity: forbidden() / unauthorized(), nearest-wins not-found, sync-or-await params (Close Next.js 16 file-routing parity gaps (async params, forbidden/unauthorized, global-error/not-found, instrumentation) #848)
• Path aliases with no build and no tsconfig paths (the # import alias) (feat: # path-alias imports + convert in-repo apps & scaffold #565)
• HTTP-verb server actions: GET/POST/PUT/DELETE with cache, tags, invalidates (feat: HTTP-verb server actions via config exports (GET/POST/PUT/PATCH/DELETE) [epic] #488)
• Cancelling server actions with AbortSignal (feat: AbortSignal cancellation for RPC, wired to async-render supersede #492)
• Component-level Suspense streaming for web components (feat: async render() + component-level Suspense (bare-await data fetch) #470/feat: <webjs-suspense> streaming SSR boundary + per-component streaming (follow-up to #469) #471)
• Eliminating the hydration re-fetch via SSR action-result seeding (feat: seed SSR action results into hydration so async render does not re-fetch (follow-up to #469) #472)
• Sanitizing production server-action errors (generic message + digest) (fix: sanitize prod server-action errors to a generic message + digest #764)
• Native SQLite on Node and Bun (dropping better-sqlite3) (feat: remove better-sqlite3, use native node:sqlite + bun:sqlite #670)
• Bun.serve vs node:http listener performance (feat: Bun.serve listener adapter behind a runtime-neutral shell seam #512, perf: cut Bun listener per-request overhead (IP clone + compress bridge) #773)
• Automatic elision: import-only pages ship zero page JS (feat: elide page/layout modules that only import components #609)
• Device-adaptive link prefetch that does not bloat the network tab (feat: device-adaptive link prefetch, viewport-safe on mobile #594)
• Per-action middleware with shared context (feat: composable per-action middleware with typed context #490)

Implementation notes (for the implementing agent)

• Where to edit: add blog/<slug>.md files at the repo root. The blog index reads this dir directly (website/modules/blog/queries/list-posts.server.ts, BLOG_DIR = <repo>/blog), sorts by date DESC, and requires title + date front matter; description, tags (comma-separated), author optional.
• Front matter shape (copy an existing post, e.g. blog/csrf-protection-without-tokens.md): title, date (ISO with +05:30), slug, description (SEO), tags, author: Vivek.
• Landmines: AGENTS.md invariant 11 bans em-dashes (U+2014), space-surrounded hyphen/semicolon as pause punctuation, and a colon after a code-shaped LHS. The .claude/hooks/block-prose-punctuation.sh hook scans NEW content and will block a commit that violates it. Write clean prose.
• No src changes, so the doc/scaffold/test commit gates do not apply; content-only.
• Invariants: keep the first-person voice, explain acronyms in parentheses for beginners, one concept per post.

Acceptance criteria

• One blog/<slug>.md per topic above, valid front matter, appears in the index sorted by date
• No em-dashes or banned pause-punctuation (hook passes)
• Dates form an even cadence with existing posts
• All posts in a single PR