agentleFS
powerczukpowerczuk/seo-machine/skills/meta-tags/SKILL.md

meta-tags

Skill0 starsChanged 2 months agoMITView on GitHub

Click-optimized meta titles, descriptions, slugs, and OG tags — five variants each, written as search-result ad copy.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before/after table when optimizing an existing page, with the reason each change should lift CTR.

    line 36
  2. Distinctive:

    Never use example brands, dates, or figures that are not from the page or the user.

    line 38

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/meta-tags — Meta element optimizer

Target: $ARGUMENTS

Why titles and descriptions matter beyond "metadata": they are inputs to click-based re-ranking (${CLAUDE_PLUGIN_ROOT}/skills/seo-fundamentals/references/navboost.md). Write them as ad copy with intent match; the first screen of the page must deliver what the title promises. Lengths and rules: ${CLAUDE_PLUGIN_ROOT}/skills/seo-fundamentals/references/scoring-spec.md.

Process

  1. Current state (URL given): python3 ${CLAUDE_PLUGIN_ROOT}/scripts/fetch_page.py "<url>" --json → current title, description, OG, canonical, H1. Measure lengths; note duplicates, missing keyword/entity, weak copy. Pull the query's GSC CTR vs position if available.
  2. Keyword and entity — confirm the primary entity and keyword, 2–3 secondary terms, intent, whether an AI Overview appears for the query (it depresses CTR regardless of copy).
  3. Titles — 5 variants, 50–60 characters, primary entity front-loaded, brand at the end if space allows:
    • A — Entity-first: "<Entity>: <benefit>"
    • B — Sub-query targeted: answers a likely fan-out question
    • C — Negative-space differentiator: "<Topic> without <common pitfall>"
    • D — Freshness-bracketed: "[<current month year>]" only when the content is genuinely updated
    • E — Number + specific promise: "<N> <things> <that do X>" Classical fallbacks: "How to <action> <without obstacle>", "<Topic> guide: <what the reader gets>".
  4. Descriptions — 5 variants, 150–160 characters, keyword included, value proposition + CTA, intent match; formulas: direct value prop · question matching intent · specific fact from the page (real, from the content) · benefit + risk reversal · curiosity.
  5. Slug — 3–5 words, lowercase-hyphenated, keyword included, stop words removed (do not change the slug of an indexed page without a 301 plan).
  6. Open Graph / Twitter — og:title (can be more descriptive), og:description, og:type, og:image 1200×630, twitter:card summary_large_image, og:locale.

Output (per page)

  • Title table: # · title · length · entity front-loaded ✓/✗ · variant · recommended.
  • Description table: # · description · length · keyword ✓ · CTA ✓ · recommended.
  • Slug recommendation.
  • OG/Twitter block as HTML <meta> tags.
  • Before/after table when optimizing an existing page, with the reason each change should lift CTR.

Never use example brands, dates, or figures that are not from the page or the user.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.