agentleFS
6t9xstar6t9xstar/one-step-seo/skills/content-quality/SKILL.md

content-quality

Skill5 starsChanged 36 days agoMITView on GitHub

Evaluate content substance, word count thresholds by page profile, answer-first openings, extractable tables and lists, FAQ sections, and semantic HTML landmarks (<main>/<article>).

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 1 is distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If C01-thin fires, identify missing subtopics, specifications, or user questions that should be answered on the page.

    line 31
  2. Very specific:

    Check whether the opening 120 words answer an informational question (what, why, how, when, where, which, who) or include an explicit …

    line 33
  3. Very specific:

    Verify the page includes at least one <ul>/<ol> or <table> (C03-extractable).

    line 35
  4. Very specific:

    Check for a visible FAQ section (C04-faq) and ensure it matches FAQPage JSON-LD (S03-faq-parity).

    line 37
  5. Distinctive:

    Never claim a page is "high quality" or "authoritative" based solely on word count.

    line 54
  6. Very specific:

    Never fabricate statistics, citations, customer quotes, or research findings when drafting content expansions or answer capsules.

    line 68

≈ “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

content-quality — Content Depth & Extractability

1. Description & Activation Criteria

Activate content-quality when the user wants to:

  • Evaluate whether a page has sufficient depth (C01-thin), an answer-first opening (C02-answer), structured lists/tables (C03-extractable, C05-table-headers), FAQ coverage (C04-faq), or semantic HTML landmarks (G11-semantic-content).
  • Calibrate content checks for specific page types (--profile blog|product|docs|home).
  • Improve both human readability and machine extractability for search snippets and AI summaries.

2. Inputs & Prerequisites

  • Target: URL, local HTML file, or build directory.
  • Profile calibrations (--profile):
    • default / blog: Thin-content threshold 200 words (C01-thin P1); checks C02-answer (P1) and C04-faq (P2). Note: the AI Visibility substantive signal (GEO_WEIGHTS.substantive, +15 pts) requires >=300 words across all profiles.
    • product: Thin-content threshold 120 words; skips C02-answer (product specs and pricing tables carry the answer).
    • docs: Thin-content threshold 200 words; elevates C04-faq and G01-llms to P1.
    • home: Thin-content threshold 100 words; skips C02-answer.

3. Step-by-Step Workflow

  1. Audit with the matching page-type profile:
    • Run npx one-step-seo page <target> --profile <blog|product|docs|home> --skills content-quality,geo-ai-search --out ./seo-report.
  2. Assess content substance (C01-thin / C01-length / C01-long):
    • Inspect report.page.wordCount (computed after stripping <head>, <script>, <style>, <noscript>, and HTML comments).
    • If C01-thin fires, identify missing subtopics, specifications, or user questions that should be answered on the page.
  3. Evaluate answer-first structure (C02-answer):
    • Check whether the opening 120 words answer an informational question (what, why, how, when, where, which, who) or include an explicit summary marker (TL;DR:, In short:, Key takeaway:).
  4. Check structured extractability (C03-extractable, C05-table-headers, C04-faq):
    • Verify the page includes at least one <ul>/<ol> or <table> (C03-extractable).
    • Confirm every <table> uses <th> header cells (C05-table-headers) so search engines and LLMs can map cell values to column names.
    • Check for a visible FAQ section (C04-faq) and ensure it matches FAQPage JSON-LD (S03-faq-parity).
  5. Check semantic document landmarks (G11-semantic-content):
    • On articles of 600+ words, confirm prose is wrapped in <main> or <article> so parsers can isolate primary content from navigation boilerplate.

4. Tools & CLI Commands

# Audit an article with the blog profile
npx one-step-seo page https://example.com/blog/guide --profile blog --skills content-quality --out ./seo-report

# Audit documentation pages with stricter FAQ requirements
npx one-step-seo audit https://docs.example.com --pages 10 --profile docs --out ./seo-report

5. Evidence Requirements

  • Cite report.page.wordCount, the active --profile, and the exact evidence from C01, C02, C03, C04, C05, and G11.
  • Never claim a page is "high quality" or "authoritative" based solely on word count; report structural readiness signals honestly.

6. Expected Outputs

  1. Content Signal Scorecard: Word count vs profile threshold, answer-first status, list/table count, table <th> status, FAQ detection, and <main>/<article> landmark presence.
  2. Structural Restructuring Plan: Drafted 40–60 word opening answer block, semantic <table> with <th> headers, or FAQ section grounded strictly in the page's existing facts.

7. Limitations & External Data

  • Automated CLI checks: Word count (body prose only), informational question + answer opening detection, <ul>/<ol>/<table> counts, <th> header presence, FAQ heading/schema detection, and <main>/<article> landmarks.
  • Advisory / manual judgment: Factual accuracy, originality, tone of voice, and subject-matter expertise require human editorial review.

8. Safety & Approval Rules

  • Never fabricate statistics, citations, customer quotes, or research findings when drafting content expansions or answer capsules. Use explicit [TODO: insert verified metric] placeholders where user data is needed.

9. Examples

# Check content quality on a product page using the product calibration
npx one-step-seo page https://example.com/shop/item --profile product --skills content-quality --out ./seo-report

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.