| name | dynamic-code-review |
| description | Use when the user asks for a deep code review, adversarial review, multi-agent review, risk-ranked review, PR review, security-sensitive review, or implementation verification with concrete file and line evidence. |
| metadata | {"short-description":"Adversarial code review workflow"} |
Dynamic Code Review
Use this skill for rigorous code review with explicit evidence and independent challenge.
Review Contract
Start by identifying:
- review scope
- intended behavior
- changed files
- critical risk areas
- available tests
- expected output format
Findings lead the response. Order them by severity and include file and line references.
Workflow
- Map the changed surface and nearby dependencies.
- Identify correctness, security, data, concurrency, API contract, and test risks.
- Use
explore for independent codebase lookup when subagent review is requested.
- Use
critic for adversarial challenge when subagent review is requested.
- Verify findings against source lines and executable evidence where practical.
- Report only actionable findings with concrete reproduction or reasoning.
Severity
- P0: data loss, security breach, production outage, broken core path
- P1: likely user-visible regression, broken API contract, missing critical validation
- P2: meaningful edge-case bug, weak test coverage around changed behavior
- P3: maintainability issue with clear future cost
Output
Return:
- findings first
- open questions
- test gaps
- brief change summary only when useful
For each finding include:
- severity
- file and line
- impact
- fix direction
- evidence
