Scored across 35 tools
Most tools are clearly separated by resource and action—import, export, QC, diversity, metadata, and server administration are distinct clusters. A few pairs (list_content vs list_variant_sets, diversity_pca vs diversity_structure) could be confused, but the descriptions provide enough context to disambiguate them.
Names are uniformly lower_snake_case and follow a predictable prefix convention: verb_resource for operations (list_*, import_*, export_*, get_*, search_*, count_*) and domain-prefixed nouns for analyses (qc_*, diversity_*). No camelCase or arbitrary synonyms appear, so the pattern is highly consistent.
With 35 tools this is a very large MCP surface, clearly above the 25+ threshold for 'too many'. Each tool is individually purposeful, but the sheer number makes selection harder and suggests the server would benefit from consolidation or splitting into focused sub-servers.
The server covers the core analytical lifecycle well: import, validate, list, filter, QC, diversity analysis, and export all have dedicated tools with progress and abort handling. Gaps are mostly administrative—no deletion/removal of databases/projects/runs and no metadata update/delete—but these are minor for the primary analysis use case.