Code Review Skill
Multi-agent code analysis with a simplicity gate, focused reviewers, and challenge validation.
Uses explicit subagent dispatch patterns from code-subagents.
Prerequisites
Arguments
Command Routing
| Invocation |
Behavior |
| (no arguments) |
Review diff to main branch |
rq |
Review diff to main branch |
rq main |
Review diff to main branch |
rq develop |
Review diff to develop branch |
feat/foo |
Review diff to feat/foo (bare branch = rq) |
rs |
Respond to review findings (interview mode) |
Subagent Architecture
Use these concrete harness subagent types. If an exact match is unavailable, use the most correct available subagent based on the harness-provided descriptions.
subagent_type |
Purpose |
sentinel |
Triage only: changed-file analysis, context retrieval, reviewer selection |
oracle |
Reviewer personas, evidence-based critique, failure-mode analysis, challenge validation |
architect |
Architecture, design-boundary, data-model, and API-contract review |
Reviewer names such as Security, Correctness, Maintainability, and PerformanceOperator are prompt personas, not subagent types. Do not use general; it is not a harness agent.
Simplicity is a mandatory reviewer persona for migrations, refactors, and architectural
changes. Run it before the other reviewers; do not include it in the parallel reviewer batch.
Review Priority
Review in this order:
- User goal and prior behavior
- Necessity and deletion
- Correctness and security
- Architecture
- SDD compliance
Treat the SDD as evidence, not as authority for whether code is necessary. A finding that
would expand behavior or infrastructure requires user approval. Never apply it as an ordinary
review fix.
rq (Request Review) Subagents
| Step |
Parallel |
Purpose |
| 1. Triage |
No |
Detect context, select reviewers, identify relevant skills to look for |
| 2. Simplicity |
No |
Mandatory necessity and deletion review for migrations, refactors, and architectural changes |
| 3. Reviewers |
Yes (per reviewer) |
Correctness, security, and other specialty analysis |
| 4. Synthesis |
No |
Deduplicate findings inline |
| 5. Architect |
No |
Architecture review |
| 6. SDD + Challenge |
No |
Check SDD compliance last, then validate findings |
rs (Respond to Review)
No subagents. Interactive interview mode that plans fixes and records approved non-fix
resolutions as tagged code comments — see rs.md.
Dispatch Patterns
Follows code-subagents patterns:
- Parallel dispatch for independent reviewers
- Sequential dispatch for dependent steps
- Fresh subagent per task — no context pollution
- Relevant skill search pre-step before each analysis phase
- Error handling: Log failures, continue with partial results
Agent Dispatch
| Agent |
Used In Step |
sentinel |
Triage only (context retrieval, file analysis) |
oracle |
Reviewers and challenge validation |
architect |
Architect (architecture review) |
Synthesis is performed inline by the main agent.
References
| Reference |
Purpose |
| rq.md |
Request review workflow - detailed steps with prompts |
| rs.md |
Respond to review workflow - interview mode |
| reviewers.md |
Reviewer definitions and prompts |
| output.md |
Output format specification |
Workflow Routing
- No arguments,
rq, or bare branch → rq.md
rs → rs.md
1---2name: code-review3description: Multi-agent code review with parallel specialized reviewers, architecture validation, challenge validation, and durable handling of previously decided findings. Use `rq` to request a review of diffs (defaults to main branch), `rs` to respond to findings and record intentional non-fix decisions beside the relevant code. Triggers on "review this", "review my code", "code review", "check for bugs", "audit this", when examining PRs, pull requests, branches, or diffs. Always asks user before applying fixes or adding decision comments.4---56# Code Review Skill78Multi-agent code analysis with a simplicity gate, focused reviewers, and challenge validation.910Uses explicit subagent dispatch patterns from [code-subagents](../code-subagents/SKILL.md).1112## Prerequisites1314- **Required**: git1516## Arguments1718### Command Routing1920| Invocation | Behavior |21|------------|----------|22| *(no arguments)* | Review diff to main branch |23| `rq` | Review diff to main branch |24| `rq main` | Review diff to main branch |25| `rq develop` | Review diff to develop branch |26| `feat/foo` | Review diff to feat/foo (bare branch = rq) |27| `rs` | Respond to review findings (interview mode) |2829## Subagent Architecture3031Use these concrete harness subagent types. If an exact match is unavailable, use the most correct available subagent based on the harness-provided descriptions.3233| `subagent_type` | Purpose |34|-----------------|---------|35| `sentinel` | Triage only: changed-file analysis, context retrieval, reviewer selection |36| `oracle` | Reviewer personas, evidence-based critique, failure-mode analysis, challenge validation |37| `architect` | Architecture, design-boundary, data-model, and API-contract review |3839Reviewer names such as `Security`, `Correctness`, `Maintainability`, and `PerformanceOperator` are prompt personas, not subagent types. Do not use `general`; it is not a harness agent.4041`Simplicity` is a mandatory reviewer persona for migrations, refactors, and architectural42changes. Run it before the other reviewers; do not include it in the parallel reviewer batch.4344## Review Priority4546Review in this order:47481. User goal and prior behavior492. Necessity and deletion503. Correctness and security514. Architecture525. SDD compliance5354Treat the SDD as evidence, not as authority for whether code is necessary. A finding that55would expand behavior or infrastructure requires user approval. Never apply it as an ordinary56review fix.5758### rq (Request Review) Subagents5960| Step | Parallel | Purpose |61|------|----------|---------|62| 1. Triage | No | Detect context, select reviewers, identify relevant skills to look for |63| 2. Simplicity | No | Mandatory necessity and deletion review for migrations, refactors, and architectural changes |64| 3. Reviewers | Yes (per reviewer) | Correctness, security, and other specialty analysis |65| 4. Synthesis | No | Deduplicate findings inline |66| 5. Architect | No | Architecture review |67| 6. SDD + Challenge | No | Check SDD compliance last, then validate findings |6869### rs (Respond to Review)7071No subagents. Interactive interview mode that plans fixes and records approved non-fix72resolutions as tagged code comments — see [rs.md](./references/rs.md).7374## Dispatch Patterns7576Follows [code-subagents](../code-subagents/SKILL.md) patterns:77- **Parallel dispatch** for independent reviewers78- **Sequential dispatch** for dependent steps79- **Fresh subagent per task** — no context pollution80- **Relevant skill search pre-step** before each analysis phase81- **Error handling**: Log failures, continue with partial results8283## Agent Dispatch8485| Agent | Used In Step |86|-------|--------------|87| `sentinel` | Triage only (context retrieval, file analysis) |88| `oracle` | Reviewers and challenge validation |89| `architect` | Architect (architecture review) |9091Synthesis is performed inline by the main agent.9293## References9495| Reference | Purpose |96|-----------|---------|97| [rq.md](./references/rq.md) | Request review workflow - detailed steps with prompts |98| [rs.md](./references/rs.md) | Respond to review workflow - interview mode |99| [reviewers.md](./references/reviewers.md) | Reviewer definitions and prompts |100| [output.md](./references/output.md) | Output format specification |101## Workflow Routing102103- No arguments, `rq`, or bare branch → [rq.md](./references/rq.md)104- `rs` → [rs.md](./references/rs.md)
martinffx