Claude SEO is a broad audit plugin rather than one narrow checklist. Its current README describes 26 sub-skills and 19 specialist agents covering technical SEO, content quality, structured data, AI-search visibility and several business-specific audit paths.
Key takeaways
- The root SEO skill routes work to more specific skills when the task is clear.
- Setup creates an isolated Python environment and installs Playwright Chromium according to the README.
- Several optional extensions connect to third-party SEO, analytics, crawling or image services.
What does Claude SEO do?
The plugin can combine crawl checks, on-page analysis, schema review, image assessment, performance evidence, international SEO and GEO. Its routing design matters because a focused schema or hreflang task should not automatically run a full audit. Recommendations still need to be separated from observed evidence, especially when live APIs or field data are unavailable.
How does the workflow work?
- Install the public plugin and run the documented setup and doctor checks.
- Select the narrowest sub-skill or command that matches the audit question.
- Gather live crawl, page, performance or platform evidence available to the environment.
- Separate observed issues from hypotheses and produce testable, prioritized remediation.
The exact result still depends on the model, the files available in the working environment, and the permissions granted to that environment. This page documents the repository instructions; Anavem did not run an end-to-end benchmark of the skill.
What does a useful first task look like?
Start with one narrow question, such as whether five product pages emit valid canonical and Product schema signals. Record the URLs and acceptance criteria, run only the relevant technical and schema checks, and retain raw evidence. Do not turn the first run into a site-wide ranking forecast.
How can you verify the result?
- Retain URLs, timestamps and raw evidence for every reported issue.
- Separate observed defects from recommendations and unavailable checks.
- Re-run the same checks after remediation and record the changed public behavior.
A successful run should produce an output you can inspect independently. If the result cannot be checked against a file, rendered artifact, source reference or explicit acceptance criterion, narrow the task before relying on it.
When is it a good fit?
- Structured technical and content audits with explicit evidence.
- Coordinating several SEO specialties across a larger site.
- Preparing implementation guidance while documenting unavailable data.
Choose it when those tasks match the documented scope. A popular repository is not evidence that a skill is appropriate for confidential or production data.
What should you review before installing it?
- The setup script, Python environment and Playwright installation.
- Credentials and data scopes for every optional connector or extension.
- Whether a recommendation is based on live evidence, a public source or an unverified assumption.
Read the current SKILL.md and every bundled script before installation. Pin a reviewed commit when repeatability matters, then test with non-sensitive data and the narrowest available permissions.
When is it not the right choice?
- Guaranteeing rankings, traffic or AI citations.
- Treating simulated scores as Search Console or analytics data.
- Running broad crawls against systems without authorization or rate controls.
Why is it included in this research set?
GitHub’s API reported 18,233 stars and 2,673 forks for AgriciDaniel/claude-seo on 2026-10-03. The public Agent Skills trend index placed it in the top ten and recorded 685 stars added over seven days.
GitHub stars and short-term growth are discovery signals, not quality scores. They do not prove security, correctness, maintained compatibility, or individual-skill usage.