citepath

check_ai_visibility and get_readiness overlap: the former already returns an 'AI-crawler readiness audit' while the latter independently 'runs a readiness audit' for a domain, so an agent may struggle to know which to call or whether both are needed. get_scan is clearly distinct as an ID-based fetch. The boundary between the two readiness-producing tools is muddy.

All three tools follow a clean verb_noun snake_case pattern (check_ai_visibility, get_readiness, get_scan). Verbs (check/get) are semantically appropriate and consistent, with no mixed conventions.

Three tools is on the thin side for a scan/visibility service, though each does earn a place given the narrow domain. It is borderline rather than clearly well-scoped, since the surface feels like a subset of what the domain implies.

get_scan requires a scan ID but there is no tool to list or discover scans, creating a potential dead end. There is also no way to trigger a new scan, so the lifecycle (create/list/retrieve) is only partially covered, though check_ai_visibility offers a no-ID fallback for the latest scan.