From brooks-agent-team
Use when completing any significant feature, fixing a non-trivial bug, or before merging — performs structured code review as the Surgeon's trusted alter ego who knows the full design intent
How this skill is triggered — by the user, by Claude, or both
Slash command
/brooks-agent-team:copilotThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
The Copilot is the Surgeon's most trusted reviewer — the person who knows the system as well as the Surgeon does, cares as much about its quality, and is specifically tasked with finding what the Surgeon missed.
The Copilot is the Surgeon's most trusted reviewer — the person who knows the system as well as the Surgeon does, cares as much about its quality, and is specifically tasked with finding what the Surgeon missed.
The Copilot is NOT a generic code reviewer. The Copilot reviews against design intent, not just code style.
Invoke inline (the Surgeon performs structured self-review using this skill) when:
Dispatch as subagent (using agents/copilot.md) when:
The Copilot reviews in this exact order. Do not skip levels.
Every finding is classified. Do not report findings without classification.
| Class | Meaning | Blocks merge? |
|---|---|---|
| Blocking | Correctness error, missing spec behavior, security issue | Yes |
| Important | Readability problem, missing test, design concern | Surgeon decides |
| Suggestion | Style, naming, optional improvement | No |
## Copilot Review: [feature/component name]
### Blocking
- [file:line] [description of issue and why it blocks]
### Important
- [file:line] [description and recommendation]
### Suggestions
- [file:line] [description]
### Summary
[One paragraph: overall assessment, confidence level, recommended next action]
The Copilot's review notes are sufficient for another Surgeon instance to continue work if the session ends. Review output should always be committed or saved — it is project documentation, not ephemeral feedback.
npx claudepluginhub zakame/skills-marketplace --plugin brooks-agent-teamProvides a checklist for code reviews covering functionality, security, performance, maintainability, tests, and quality. Use for pull requests, audits, team standards, and developer training.