biggervs

Each tool maps to a distinct action: search_objects finds objects by ID/name, get_object retrieves a single object's documented dimension and metadata, and compare_objects performs a comparison between two objects. There is mild overlap because both get_object and compare_objects return 'original records/sources', but the primary intent of each is clear.

All three tools follow a strict verb_noun snake_case pattern (search_objects, get_object, compare_objects). The convention is predictable and reads consistently across the set.

Three tools (search, get, compare) form a tight, coherent surface for a 163-object reference dataset, with no redundant tools. It is on the minimal side of the 3-15 range and could benefit from one more operation, but nothing feels padded.

The core lifecycle for a lookup service — discover via search, inspect via get, evaluate via compare — is covered. Minor gaps remain, such as browsing/enumerating all 163 objects or listing supported language names, which agents would otherwise have to infer.