SEO skill generator
Enter a URL. We fetch the page, robots.txt, the sitemap, host variants and a crawler diff, then write a SKILL.md that names what we actually saw. Install it so an agent checks this host instead of a generic SEO checklist.
Up to 12 fetches: the page, robots.txt, the sitemap (and first child if it is an index), three host variants, a four-persona crawler diff, and TLS. Same 60/hour budget as the other tools.
Illustrative result for a reserved example URL. Your check replaces it.
Nothing blocking
Snapshot 2026-09-03T12:00:00Z. 12 fetches used. SKILL.md for example.com.
- HTTP
- 200
- Indexable
- yes
- Title
- Example Domain
- robots.txt
- Googlebot: allowed
- Sitemap
- 1 URL
- Fetches used
- 12
This is what a clean snapshot looks like. Enter a URL above to get the same report for any site.
What this check means
- Skill file
- A markdown file with YAML frontmatter that an agent loads as instructions. This one is bound to one host and one measurement time.
- Measured, not advised
- The file records status, indexability, robots, canonical, title, H1 and sitemap count from this check. It does not invent rankings.
How to read the result
- URL
- The page to measure. Hostname is taken from the final URL.
- robots.txt
- Fetched from the origin. Missing means default-allow.
- Sitemap
- The Sitemap: directive if present, otherwise /sitemap.xml.
Common causes
- A staging noindex that should not be in the skill as a permanent fact — re-generate after the fix.
- A sitemap 404 because the host only publishes sitemap-index.xml.
- A JavaScript-only title the HTTP fetch cannot see.
What does the generator actually fetch?
Up to 12 fetches: the page, robots.txt, the sitemap (and the first child if it is a sitemap index), three host/scheme variants, a four-persona crawler diff (Googlebot, browser, Bingbot, GPTBot), and the TLS handshake.
How is this different from running the individual tools?
The individual tools report one check each. The SKILL.md combines all of them into one file with a machine-readable expected-state block and a per-field runbook, so an agent can diff a post-deploy re-check against the snapshot.
Where do I put SKILL.md?
In the skills folder your agent already reads (for example .agents/skills/ or a Claude skill directory). The YAML name field is seo-<host>.
Is this a rank report?
No. The generator fetches the page, robots.txt, a sitemap, host variants and a crawler diff over HTTP. It does not render JavaScript and it does not look up rankings.