Your objective is to implement $adversary audit.
Verify that adversary reviewers remained read-only, did not approve their own work, and produced evidence-linked findings. Report any reviewer lockout or critic separation violation.
Anti-Rationalization
| Shortcut / Temptation | Why It Fails |
|---|---|
| Treat the skill as complete because the file exists | Skill contracts must describe observable behavior and verification, not just command names. |
| Skip artifact and safety checks for a small command | Small commands still mutate state or guide execution; preserve the same gates. |
Exit Criteria
- Command behavior matches the owning feature contract.
- Required reads, writes, and user-visible outputs are described.
- Safety, governance, and artifact constraints are preserved.
- Missing state produces a clear, non-destructive result.