メニュー
Verify a refactor preserved behavior — composes contract, side-effect, and CFG-equivalence checks across two versions of a function. Use after editing a function to confirm the change is safe.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Audit whether an AI agent edit to a file is safe — composes secret-surface, taint-flow, change-impact, and API-surface analysis into an agent-edit safety report. Use before letting an agent modify a sensitive or high-blast-radius file.
Execute structural decompositions based on an architectural analysis report. Use when breaking circular dependencies, splitting god classes, reducing coupling between modules, extracting interfaces at seam boundaries, or executing any large-scale structural change identified in an architectural analysis. Requires a prior architecture audit (the architecture-audit skill) that produced project-map.md and a run report.
Use when asked to audit architecture, produce a comprehensive architectural overview, get the full structural picture of a codebase, assess overall health and porting readiness, find module boundaries, circular dependencies, coupling hotspots, dead code, or code patterns. Depth 3 — full audit with hypothesis-driven investigation. Optionally enriches the structural audit with runtime and git evidence (coverage, churn, co-change coupling, ownership) when those overlays are available. Replaces the architectural-analysis and architecture-audit-plus skills.
Use when looking for extraction candidates, planning a module split, analyzing module boundaries, or before decomposing a large component. Also use when asked "where should I split this?" or "find natural module boundaries". Depth 2 — investigate. Produces extraction candidates with cut costs, layer violations, interface width assessment, orphan types, and a recommended decomposition sequence.
Use before modifying a file or symbol, or when asked "what breaks if I change X?". Depth 0 — fast, no follow-up queries, returns immediately. Two modes: inline (no artifacts, default for agent-initiated use) and artifact (writes to docs/act/, triggered for human-initiated requests or Critical risk verdicts).
Traverse large repositories efficiently using act's query tools. Use when exploring unfamiliar code, mapping dependencies, understanding API surfaces, analyzing side effects before modifying functions, or following call chains across files. Avoids reading entire files by querying only the structure needed.
| name | verify-refactor |
| description | Verify a refactor preserved behavior — composes contract, side-effect, and CFG-equivalence checks across two versions of a function. Use after editing a function to confirm the change is safe. |
Confirm a function's refactor is behavior-preserving by composing three
Engineer-tier verification ops over a two-version comparison (git working-tree
vs HEAD by default, or an explicit before/after pair).
Scope note: this skill verifies ONE function pair. For a changed SET of
functions with git context, use act gate (see the safe-to-merge skill) — its
verdict vocabulary (MERGE/REVIEW/BLOCK/UNKNOWN) is intentionally distinct from
this skill's per-function SAFE/REVIEW/UNSAFE/UNKNOWN.
| Step | Tool | What it answers |
|---|---|---|
| 1 | verify_contract_preserved | Did the public + behavioral contract (signature, effects, control, returns, guards, raises) change? |
| 2 | verify_side_effects | Which side effects were added/removed? Was a cleanup dropped? |
| 3 | verify_behavioral_equivalence | Is the control-flow shape equivalent (no new/removed branches, loops, exception paths)? |
All three are Engineer. verify_behavioral_equivalence at scope:"port" is Enterprise.
verify_contract_preserved with target + file. If the verdict is
broken, report the broken dimensions and stop — the refactor changed the
contract.verify_side_effects. Surface any dropped_cleanup: true prominently —
a removed write/close while an allocation is kept is a likely leak.verify_behavioral_equivalence. equivalent confirms the shape held;
changed lists which dimensions; unknown means a grammar dimension (e.g.
Go exceptions) could not be judged — say so, do not claim safety.changed in an
expected way; summarize what moved.broken, dropped_cleanup: true, or behavior changed
unexpectedly.unknown on a not-modeled dimension; never
upgrade UNKNOWN to SAFE.Report: the verdict, the three op results in one line each, and the specific
dimensions/effects that changed. Quote modeled_kinds when a verdict is UNKNOWN.