No user-submitted related servers found.
Scored across 10 tools
Every tool has an explicit 'Use when' / 'Don't use when' clause that names the alternative tool to use instead, e.g. scan_repo vs get_chunk, write_fix vs git_commit, read_memory vs write_memory, score_risk vs get_health_score. Boundaries are crisp even for near-neighbors like infer_architecture (which writes state) and read_memory (which only reads it).
Names overwhelmingly follow a snake_case verb_noun pattern (scan_repo, get_chunk, write_fix, read_memory, score_risk, generate_changelog). The only minor deviation is git_commit, which uses a resource prefix rather than a leading verb, but it remains readable and predictable.
Ten tools is well within the ideal range and each maps to a distinct step in a repo-cleanup workflow (scan, inspect, patch, commit, persist state, infer, score, report). No tool feels redundant or padded.
The surface covers the full lifecycle: scan_repo discovers findings, get_chunk inspects, write_fix patches, git_commit records, memory tools persist, and scoring/reporting tools close the loop. Minor gaps exist — no standalone test-runner or revert/undo tool, and tests only run bundled inside write_fix — but agents can work around these.