agentleFS
nguyenthanhtatnguyenthanhtat/screen1-claude/.claude/skills/check-seo/SKILL.md

check-seo

Skill2 starsChanged 26 days agoMITView on GitHub

Audit any website's SEO, suggest prioritized fixes, then auto-apply fixes to the local codebase. Check and install if missing: For Core Web Vitals (optional): Run the audit with LLM-optimized output: Parse the XML output to extract: - Overall score (0-100) and grade (A-F) - Per-category scores and weights - All issues with severity: critical, warning, info Display a summary table: Group all issues by severity and show the top 10 prioritized by impact: Ask the user: "Found X critical issues…

  • Installs packages

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Check and install if missing:

    line 10
  2. Distinctive:

    Run the audit with LLM-optimized output:

    line 22
  3. Very specific:

    All issues with severity: critical, warning, info

    line 30
  4. Distinctive:

    If "select", list issues numbered so user can pick which to fix.

    line 62
  5. Distinctive:

    Never fix what cannot be confidently mapped to a file

    line 99
  6. Distinctive:

    Always show before/after for each edit

    line 100
  7. Distinctive:

    If a file doesn't exist and needs to be created, create it

    line 101

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

check-seo

Audit any website's SEO, suggest prioritized fixes, then auto-apply fixes to the local codebase.

When to use

Triggered by: "check SEO", "audit SEO", "fix SEO", "my SEO score", "audit https://...", "improve SEO"

Prerequisites

Check and install if missing:

seomator --version 2>/dev/null || npm install -g @seomator/seo-audit

For Core Web Vitals (optional):

npx playwright install chromium

Phase 1 — Audit

Run the audit with LLM-optimized output:

seomator audit <url> --format llm

Parse the XML output to extract:

  • Overall score (0-100) and grade (A-F)
  • Per-category scores and weights
  • All issues with severity: critical, warning, info

Display a summary table:

Overall: 67/100 (D)

Category         Score  Issues
Core SEO          45    3 critical, 2 warnings
Performance       72    1 critical, 4 warnings
Security          88    0 critical, 1 warning
...

Phase 2 — Suggest

Group all issues by severity and show the top 10 prioritized by impact:

CRITICAL (fix these first):
1. [Core SEO] Missing meta description — affects click-through rate
2. [Core SEO] Title tag too long (87 chars, max 60) — truncated in SERPs
3. [Performance] No image lazy loading — hurts LCP score
...

WARNINGS:
4. [Security] Missing X-Frame-Options header
...

Ask the user:

"Found X critical issues and Y warnings. Fix all critical issues automatically? (yes/no/select)"

If "select", list issues numbered so user can pick which to fix.

Phase 3 — Detect Stack

From the project root, detect the tech stack:

# Check for framework config files
ls next.config.* 2>/dev/null && echo "nextjs"
ls nuxt.config.* 2>/dev/null && echo "nuxt"
ls vite.config.* 2>/dev/null && echo "vite"
find . -name "*.html" -maxdepth 3 | head -5

Detection rules:

  • next.config.* OR src/app/layout.tsx OR pages/_app.tsx → Next.js
  • nuxt.config.* → Nuxt
  • vite.config.* + index.html → Vue SPA
  • *.html files, no framework → Plain HTML
  • Unknown → treat as Plain HTML

Load the appropriate fix map from references/fix-maps.md.

Phase 4 — Fix

For each approved issue:

  1. Find the target file using the fix map (see references/fix-maps.md)
  2. Show the planned edit:
    File: src/app/layout.tsx
    Adding: description to metadata export
  3. Apply the fix with Edit tool
  4. Confirm the fix was applied correctly

Rules:

  • Never fix what cannot be confidently mapped to a file
  • Always show before/after for each edit
  • If a file doesn't exist and needs to be created, create it
  • Group unmappable issues into a "Manual Fixes" section at the end

Phase 5 — Verify

Re-run the audit after all fixes:

seomator audit <url> --format llm

Show score delta:

Before: 67/100 (D)
After:  84/100 (B)  ▲ +17 points

Fixed: 8 issues
Remaining manual fixes: 3 issues (see below)

Manual Fixes Report

For issues that couldn't be auto-fixed, provide specific instructions:

MANUAL FIXES REQUIRED:
1. [Performance] Enable server-side compression (gzip/brotli)
   → Configure in your hosting provider or server (nginx/vercel.json)
   → Expected impact: +5-8 points

2. [Crawlability] Submit sitemap to Google Search Console
   → https://search.google.com/search-console

Reference Files

  • references/seomator.md — CLI flags, XML output schema, scoring system
  • references/fix-maps.md — Issue-to-file mapping for each stack

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.