aihall

The three content_* tools are differentiated mainly by granularity (full site / single section / condensed summary), which is understandable but still overlapping since all return site text plus a change list. get_business_info substantially duplicates content_summary (hours, services, contact), so an agent could easily pick the wrong one for business details.

Three tools share a predictable content_<granularity> prefix (full, section, summary), which is clean, but get_business_info breaks the pattern with a different verb_noun style and no shared prefix. The mix is still readable but not a single consistent convention.

Four tools is a lean, well-scoped surface for a website-content server, with each tool covering a distinct granularity level. A dedicated lookup/search tool would round it out, but nothing is clearly redundant enough to inflate the count.

Read access is covered at three levels plus a business-metadata endpoint, and pagination is handled via nextCursor/nextOffset. However, there is no search or query operation (e.g. find a page/keyword), which forces agents to pull full content before locating anything, and the workflow depends entirely on IDs from content_full or content_summary.