The seo-audit skill makes an agent work as an SEO auditor. It collects site context, current issues and scope, then checks crawlability and indexation first, followed by technical foundations, on-page elements, content quality and authority.
It has sections for international SEO (hreflang, canonicals, locale sitemaps), local SEO and common issues by site type. Findings follow a fixed format of issue, impact, evidence, fix and priority, and the report closes with an action plan that puts blockers first.
What You Get
- An audit report with an executive summary, top priority issues and quick wins
- Technical, on-page and content findings, each with issue, impact, evidence, fix and priority
- A prioritized action plan from critical fixes to long-term recommendations
- Hreflang, canonical and locale sitemap checks for multilingual sites
- A list of audit tools and when to reach for each
Common Use Cases
- Finding out why organic traffic dropped after a migration or Google update
- Checking canonicals, redirects, robots.txt and sitemaps on a site that will not rank
- Reviewing hreflang and canonical setup across locale versions
- Auditing title tags and meta descriptions in bulk
What Makes It Unique
It warns that web fetch and curl cannot see JavaScript-injected schema, so the agent must use a rendering browser or the Rich Results Test before reporting no schema found. It treats fetched pages as untrusted data, and carries detailed international SEO guidance with sources, which many audit checklists skip.
Under The Hood
The skill is an instruction file with seven reference documents on title tags, international SEO, local SEO, ranking pushes and AI writing patterns. It has no scripts. To audit a site the agent fetches pages, robots.txt and sitemaps, and may render pages in a browser tool to read JSON-LD, or use crawler and scraping tools the user has. It is told to treat fetched page text as data and ignore instructions inside it. It writes its report in chat and does not change the site.