Scored across 35 tools
Each tool has a clearly distinct resource and action: per-type create/update tools, generic read/list/search/delete tools, and specialized tools such as get_person_tree, get_relations, get_timeline, find_duplicates, merge_objects, undo_change, and read_media. The few potentially overlapping areas, especially create_records versus individual create_* tools and search versus list_objects, are explicitly differentiated in the descriptions.
The set overwhelmingly follows a snake_case verb_noun pattern such as create_person, update_family, get_object, list_objects, and delete_object. The only notable deviation is the single-verb tool search, which is still readable and conventional.
Thirty-five tools is heavy for a single MCP server and exceeds the typical well-scoped range, even though the genealogy domain is broad and has many record types. Per-type create/update tools plus generic utilities make each tool defensible individually, but the overall set is still too large for easy agent use.
The surface covers full lifecycle operations across people, families, events, places, sources, citations, repositories, notes, media, and tags, with generic get/list/search/delete covering all types. It also includes batch creation, duplicate detection, merging, undo, recent changes, bookmarks, relationship/tree/timeline views, and media reading, leaving no obvious dead ends.