Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: audit the provided crawl data end-to-end into priority-ordered findings with traffic-impact estimates, and return the report with the fix-PR plan and the verification checklist.
- analyze — Run pure deterministic analysis over the provided crawl data with no model call and return scored, traffic-impact-estimated findings.
- check_core_web_vitals — Return only the Core Web Vitals findings — LCP, INP and CLS — scored against the analyzer's thresholds with their traffic impact.
- check_indexing — Return only the indexing-error findings: URLs excluded from the index or erroring, with the reason.
- check_broken_links — Return only the broken-link findings — the 4xx and 5xx responses in your crawl.
- check_canonical — Return only the canonical-conflict findings.
- check_hreflang — Return only the hreflang findings — the international targeting issues.
- check_schema — Return only the schema.org structured-data findings.
- plan_fix_pr — Return the code-fix PR plan — branch, title and one item per non-ok finding naming the file and the change. Always marked as requiring approval; the agent never opens or merges it.
- build_verification_checklist — Return the post-fix verification checklist: the generic steps plus one targeted step per issue category present in your data.
- list_capabilities — List the agent's static capabilities: the available tools, the SEO issue categories and the Core Web Vitals thresholds. Useful for discovery.
- plan_inputs — Plan the inputs for a tool: returns the clarifying questions, the JSON schema and a ready-to-edit example.
- discover_intent — Understand your goal and co-design the exact input through clarifying questions before running.
- open_form — Opens the guided form to collect the site and your crawl data. The entry point for running the agent on your own audit.
- run_form — Internal: invoked by the guided form when the user submits. Not for direct use.
