PACE review stage 执行 skill (Claude Code). reviewer 与 spec-compliance 并行只读返回, 主 agent 合并 passN.md 后再运行 evaluator, 最终由主 agent 落盘 VERDICT 与 next_action.
Scanned 9/5/2026
Install to Claude Code
npx -y skills add WenJunDuan/Rlues --skill athena-review --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Athena Review?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/wenjunduan-athena-review-99caf420)More formats (shields.io, HTML) on the badges page.
---
name: athena-review
description: PACE review stage 执行 skill (Claude Code). reviewer 与 spec-compliance 并行只读返回, 主 agent 合并 passN.md 后再运行 evaluator, 最终由主 agent 落盘 VERDICT 与 next_action.
---
# /athena-review — Review stage (v9.9.1)
## 触发
impl 完成、checklist 全绿、测试通过; Refactor/System 还须先完成 runtime-verify.
## 工作流 (2 + 1)
1. 主 agent 用 CC 当前 subagent 机制并行运行 reviewer 与 spec-compliance.
2. 两个 read-only agent 只返回完整 markdown 段, 不创建或修改文件.
3. 主 agent 收齐结果, 串行写入 `sprints/{current_sprint_slug}/reviews/passN.md`.
4. 主 agent 再运行 evaluator; evaluator 读取已合并的 passN.md、design、checklist 与 evidence, 返回 Evidence Cross-Check + VERDICT.
5. 主 agent 追加 evaluator 结果并更新 `_index.next_action`.
## VERDICT
| VERDICT | 下一步 |
|---|---|
| PASS | polish (Refactor/System) 或 ship |
| CONCERNS | 修复或明确 defer 后生成 passN+1; 不得直接 ship |
| REWORK / FAIL | 回 impl, 修复后生成 passN+1 |
## 门禁
ship 时 delivery-gate 选择数字最大的 passN.md, 只接受最终 PASS,并检查 Spec Compliance、Evidence Cross-Check (Refactor/System) 与 reviewer/evaluator 结果. agent 启动记录不能代替返回结果或落盘产物.
## 禁止
- reviewer/spec-compliance/evaluator 并发写同一文件
- evaluator 早于前两份结果合并
- read-only agent 更新 `_index.md`
- 主 agent 伪造 agent findings; 主 agent 只负责合并、去重与落盘
## 例外
- Hotfix: 可跳过 review
- Bugfix / Quick: 仅 reviewer
- Feature / Refactor / System: 完整 2+1 流程
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Ultra-compressed commit message generator. Cuts noise from commit messages while preserving intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why" isn't obvious. Use when user says "write a commit", "commit message", "generate commit", "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.
Ultra-compressed code review comments. Cuts noise from PR feedback while preserving the actionable signal. Each comment is one line: location, problem, fix. Use when user says "review this PR", "code review", "review the diff", "/review", or invokes /caveman-review. Auto-triggers when reviewing pull requests.
一个全面的 Claude Code 会话验证系统。
Verification loop for Django projects: migrations, linting, tests with coverage, security scans, and deployment readiness checks before release or PR.
Verification loop for Spring Boot projects: build, static analysis, tests with coverage, security scans, and diff review before release or PR.