Bowling Data Agent API

The three tools map to distinct stages of a retrieval workflow: about (catalog/metadata), search (discovery), and fetch (retrieval by id). The only mild overlap is that both about and search can surface the free sample day, but the descriptions clarify that about enumerates documents while search returns ranked passages.

All three names are single, lowercase verbs (about, fetch, search) with no mixing of conventions. Although 'about' is a noun-ish label rather than a strict verb_noun pattern, the set is uniformly terse and predictable.

Three tools is on the thin side for a data-access API, but the domain is narrow (read-only access to published bowling-center documents), so each tool earns its place. It sits at the borderline where a bit more surface, such as listing available document ids or sites, could reduce dependence on 'about'.

The discover-then-fetch lifecycle is fully covered, including cross-site search and a documented sample-day path, with no dead ends for a read-only corpus. The main gap is that there is no dedicated way to list or filter documents by site or id without first calling about or search.