No user-submitted related servers found.
Scored across 50 tools
Most tools have a clearly stated role and descriptions explicitly state routing preconditions ('use after sumo_qa_deciding_approach routes here') and contrasts ('lighter-weight than sumo_qa_creating_test_plan'). A few pairs still overlap: sumo_qa_load_catalogue_entry versus the four zero-argument loaders, sumo_qa_finding_test_data (router) versus sumo_qa_find_test_data (leaf), and sumo_qa_measuring_coverage versus sumo_qa_record_coverage — but each clash is disambiguated in prose.
The bulk share a `sumo_qa_` prefix, but two conventions coexist inside it: gerund skill names (planning_qa_rollout, creating_test_plan, strategising, triaging_test_failures) and verb_object utility names (find_test_data, scan_repo, analyze_diff_impact). The entry router `using_sumo_qa` also breaks the prefix pattern entirely, and descriptions refer to hyphenated skill ids that differ from the tool names.
50 tools is heavy for any single server and sits at the top of the 'too many' band, even though the surface is organized into routing, skill-loading, test-data, external-skill and artefact families. Several skill entry points could arguably be consolidated into the router rather than exposed as separate tools.
The QA lifecycle is covered end to end: routing, approach selection, planning/execution rollout, TDD and test strengthening, triage, coverage and mutation measurement, risk ledger and scorecard, report generation, test-data explain/find/validate/register, catalogue and knowledge ingestion, plus a full external-skill search/preview/install/execute/rollback loop. No obvious dead ends for the stated domain.