codeclone-review

name	codeclone-review
description	Review a Python repository via CodeClone MCP — first pass, baseline triage, changed-files review, and the findings that trip the gates.

CodeClone Review

Structural / clone / changed-scope / gate review. Read-only; never mutates state.

Rules

• MCP tools only (CodeClone plugin). Absolute root. CodeClone is the source of truth — never reinterpret, suppress, or mutate.
• No latest run → analyze_repository / analyze_changed_paths first.
• Default / pyproject thresholds on the first pass. Lower thresholds = explicit exploratory follow-up, never silent.
• Production- and changed-files-first over broad listing. detail_level for lists: summary | normal | full.

Workflows

• Full: analyze_repository → get_production_triage → list_hotspots → get_finding → get_remediation
• PR: analyze_changed_paths → get_report_section(section="changed") → list_findings(changed_paths=…, sort_by="priority") → generate_pr_summary
• Metrics / coverage: get_report_section(section="metrics") (coverage join → help(topic="coverage"))
• Deeper: help(topic="analysis_profile") → re-analyze with lower thresholds → compare_runs

Gates → the findings that trip them

• evaluate_gates(run_id, fail_on_new=, fail_complexity=, fail_coupling=, fail_dead_code=, fail_health=, …) → gate decision.
• See the findings behind a reasons[] token: list_findings(novelty="new", family="clones"|"complexity"|…, source_kind="production").
• Per-family, new vs known: check_clones | check_complexity | check_coupling | check_cohesion | check_dead_code.
• Drill one: get_finding(finding_id) → get_remediation(finding_id).
• Review loop: mark_finding_reviewed(finding_id) → list_reviewed_findings; exclude_reviewed=true in long sessions.

Reading the response

Key / easily-misread fields; the real response carries more.

Field	Meaning
health.score/grade	0–100 / A–F; dimensions = per-family scores
findings.new/known	baseline-relative novelty — NOT patch-local proof (use change-control verify)
new_by_source_kind	new split prod / tests / fixtures (the gate counts production)
evaluate_gates.would_fail + reasons[]	gate verdict + cause tokens (clone:new, health, …)
finding severity vs priority	severity = impact class; priority = ranked action order
finding source_kind	production / tests / fixtures — filter test noise
novelty="known"	in baseline, NOT "safe" — a patch may reintroduce it

Non-goals

• No auto-suppress, no mutate. overloaded_modules = report-only context, not a finding/gate.
• A clean default pass ≠ proof no finer-grained issues exist. Do not fall back to CLI / local report files.