MyMCPTools

Most tools target clearly distinct resources: catalog stats, a single server entry, probe history, drift events, incidents, categories, and search. The one real overlap is get_server_status, which doubles as a catalog-wide verdict summary and therefore partially duplicates get_catalog_stats; get_server_history vs list_server_incidents could also be momentarily confused but descriptions separate uptime trends from discrete outages.

Every name follows a strict verb_noun snake_case pattern (get_, list_, search_ prefixes) with nouns that consistently refer to catalog entities (server, catalog, categories, incidents, schema_drift). No casing or verb-style deviations.

Eight tools is well within the sweet spot and each covers a distinct read-only query axis (aggregate stats, single entity, history, live status, taxonomy, drift, incidents, search). Nothing feels redundant or padded.

The catalog-browsing domain is well covered end to end: discovery (search, categories), lookup (get_mcp_server), health (stats, status, history), and change signals (drift, incidents). Minor gaps remain, e.g. no way to compare servers or pull tool-level details for a specific exposed tool other than via drift events, but the core lifecycle is closed.