菜单
Fast production-first triage — health, top hotspots, baseline-relative regressions, and the recommended next review action.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Human repository setup readiness — status, plan, apply, and wizard over the CLI (not MCP).
Rank demonstrated architectural problems from one canonical run plus targeted source verification. Read-only; priorities are response-local heuristics, never CodeClone findings.
Inspect structural blast radius before editing — dependents, clone cohort, risk signals, do-not-touch boundaries. Read-only; does not declare intent.
MANDATORY HARD GATE when CodeClone MCP is connected. Before ANY repository file write, obtain edit_allowed=true from start_controlled_change. Before claiming done/verified/ready, finish_controlled_change must accept and clear the intent.
Use CodeClone Engineering Memory through MCP to retrieve scoped evidence before edits and preserve durable incidents, decisions, risks, contradictions, and verification anchors as draft candidates. Memory never grants edit permission; agents propose, humans govern.
Before broad rg/grep, non-trivial planning, unclear scope, contract/schema work, or resuming another agent — call get_implementation_context once to bound the implementation frontier. Then use targeted source reads. Read-only; never grants edit permission.
| name | codeclone-production-triage |
| description | Fast production-first triage — health, top hotspots, baseline-relative regressions, and the recommended next review action. |
Fast, production-focused first pass: health, top hotspots, baseline-relative regressions, next action — without a full review loop.
codeclone-hotspots instead.new/known is NOT patch-local proof; patch-local regressions need the change-control before→after
verify.analyze_repository(root=<abs>) → get_production_triage
Two calls. Stop unless asked for more.
get_finding(finding_id) → get_remediation(finding_id).analyze_changed_paths(root=..., paths=[...]) → get_report_section(section="changed").evaluate_gates(run_id, …) →
list_findings(novelty="new", source_kind="production", family=…).Key / easily-misread fields; the real response carries more.
| Field | Meaning |
|---|---|
health.score/grade | 0–100 / A–F |
findings.total / production / new | all / production-only / new since baseline |
new_by_source_kind | new split prod / tests / fixtures (the gate counts production) |
baseline.status | ok / mismatch / missing |
hotspot severity / source_kind | impact class / production vs tests |
novelty | baseline-relative, NOT patch-local |
root. No latest run → analyze_repository first.codeclone-review.Health + grade → counts (total / production / new) → top hotspots (≤5: path, kind, severity) → baseline status → recommended next action. Concise — a snapshot.
codeclone-review) or modify files / declare intent (use
codeclone-change-control).