Seo Audit

Comprehensive SEO audit and optimization for any website or URL. Use when asked to audit a site's SEO, analyze on-page optimization, check technical SEO issu...

Security Scan

The skill's files and runtime instructions are consistent with an SEO auditing tool; there is no evidence of obfuscated or unrelated behavior, but it fetches arbitrary URLs (SSRF/network fetch risk) and relies on platform web_search/web_fetch capabilities, which you should review before use.

Purpose & Capability

Name/description match the provided artifacts: SKILL.md describes an SEO audit workflow and included references and a Python audit script implement content, heading, meta, schema, image-alt and scoring checks. Required env vars, binaries, and config paths are none — proportional to the stated purpose.

Instruction Scope

SKILL.md confines actions to fetching pages, analyzing HTML, and using platform web_search/web_fetch. It does not instruct reading unrelated local files or environment variables. Caution: the workflow explicitly fetches arbitrary target and competitor URLs — this is expected for an auditor but introduces network-side risk (possible SSRF / access to internal resources depending on where the agent runs).

Install Mechanism

No install spec present; the skill is instruction-first and ships a small Python script for optional local/bulk use. No external downloads, package installs, or archive extraction are used.

Credentials

No credentials, secrets, or config paths are requested. The skill's needs (none) align with its functionality. Note: SKILL.md references using platform web_search/web_fetch — those platform tools may themselves require credentials or have privacy implications depending on the execution environment; the skill does not ask for any unrelated credentials.

Persistence & Privilege

always:false and normal invocation flags. The skill does not request persistent or elevated privileges and does not modify other skills or global agent settings.

Assessment

This skill appears to do what it says: fetch pages and produce an SEO audit. Before installing, consider: (1) Network fetch risk — the skill (and its script) will retrieve arbitrary URLs you or the agent provide. If the agent runs inside an environment with access to internal services, a malicious or mistaken URL could expose internal endpoints (SSRF). Only allow URLs you trust or run the skill in a network-isolated sandbox. (2) Search usage — SKILL.md expects platform web_search/web_fetch; verify what search provider will be used and whether that provider requires API keys or logs queries you consider sensitive. (3) Data handling — audit reports may include page content; avoid sending private site content to untrusted third parties. (4) The provided python script performs only HTTP fetches and HTML parsing; you can review/execute it locally to validate behavior before enabling the skill in production. If you want a stricter assessment, provide the full, untruncated seo_audit.py or details about the platform web_search/web_fetch implementations and where the agent will run (network access policy).

SEO Audit Skill

Deliver a thorough, actionable SEO audit for any URL or domain. Output should be immediately useful — not generic advice.

Workflow

1. Gather Target

Ask for the URL if not provided. Optionally ask for: target keywords, competitors to benchmark against, and priority pages.

2. Fetch & Analyze Pages

Use web_fetch to retrieve page content. Analyze:

• Title tag: Present? Length (50–60 chars ideal)? Contains primary keyword?
• Meta description: Present? Length (150–160 chars ideal)? Compelling CTA?
• H1/H2/H3 structure: Single H1? Keyword-rich headings? Logical hierarchy?
• Content quality: Word count, keyword density (1–2%), readability, duplicate content signals
• Internal linking: Anchor text quality, orphan pages, link depth
• Image alt text: Present on all images?
• URL structure: Clean, keyword-rich, no excessive parameters?
• Schema markup: Present? Appropriate type?

3. Technical Signals (infer from page source)

• Page load indicators (inline CSS vs external, render-blocking scripts)
• Mobile viewport meta tag
• Canonical tags
• robots meta tag
• HTTPS
• Structured data (JSON-LD)

4. Keyword Analysis

Use web_search to:

• Identify what the page currently ranks for (search site:domain.com)
• Find keyword gaps vs competitors
• Suggest 5–10 target keywords with estimated intent (informational/transactional/navigational)

5. Competitor Benchmarking (if competitors provided)

Fetch top competitor pages for same keywords. Compare: title structure, content length, heading strategy, backlink signals (visible in SERPs).

6. Generate Report

Use the report template in references/report-template.md. Sections:

• Executive Summary (3–5 bullet critical issues)
• Score card (0–100 across 5 dimensions)
• On-Page Analysis
• Technical SEO
• Keyword Opportunities
• Quick Wins (fix in <1 hour)
• Long-Term Roadmap

7. Run the Audit Script (optional, for bulk analysis)

python3 scripts/seo_audit.py --url <URL> [--competitors url1,url2] [--keywords kw1,kw2]

See references/scoring-rubric.md for how scores are calculated. See references/checklist.md for the full 50-point audit checklist.