AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Workflows Review

skill-jerrylalala-compound-engineering-workflows-review · by Jerrylalala

对代码、分支或计划进行多角度结构化审查,输出 findings、风险与建议

No reviews yet
0 installs
0 views
view→install

Install

$ agentstack add skill-jerrylalala-compound-engineering-workflows-review

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-jerrylalala-compound-engineering-workflows-review)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Workflows Review? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

workflows-review

用于在 Codex 中做审查,而不是执行实现。

这是 Codex 中的 review 主入口。目标不是简单写几条意见,而是做出可信、可排序、可继续处理的结构化审查。

适用对象

  • 当前分支改动
  • 指定 PR / 分支
  • 指定 plan 文档

输出原则

  • findings first
  • 按严重性排序
  • 明确区分已验证问题与假设
  • 尽量给出文件引用

审查步骤

  1. 明确审查目标:
  • 当前分支
  • 指定 PR / 分支
  • plan 文档
  1. 查看 diff、相关文件和上下文。
  2. 从这些角度审查:
  • 正确性
  • 风险与回归
  • 复杂度
  • 性能
  • 安全
  • 与现有模式的一致性
  1. 如果是 plan 审查,还要检查:
  • 是否可执行
  • 是否原子化
  • 是否缺少验证步骤
  • 是否缺少关键风险说明

高价值能力

1. 多角度审查

至少从这些角度扫描:

  • 正确性
  • 风险与回归
  • 架构一致性
  • 性能
  • 安全
  • 简化空间

如果范围较大,可以显式调用相关专长技能或子代理,但最终必须统一整合成一份 review 结果。

2. Findings 优先

默认以 findings 开头,而不是先写总结。

每条 finding 尽量包含:

  • 严重性
  • 问题描述
  • 影响
  • 文件或位置
  • 为什么这是真问题

3. 保护工件

不得建议删除:

  • docs/plans/*.md
  • docs/solutions/*.md

如果某个审查思路会导向“这些文档应删除”,要主动丢弃该建议。

4. Plan 审查专用检查

如果目标是 plan,还要检查:

  • 是否能交给 Claude 执行
  • 是否缺少 checkbox
  • 是否缺少 ## Overview
  • 是否缺少 Goal / Tech Stack
  • 是否把任务写得过大
  • 是否缺少具体验证

5. 输出可信度

不要把猜测写成结论。

如果某点只是怀疑,明确写成:

  • assumption
  • open question
  • residual risk

输出格式

优先输出:

## Findings

1. [severity] 问题描述
2. ...

然后可选输出:

  • open questions
  • residual risks
  • next step

如果没有明确问题,也要明确写:

  • no material findings

并补充剩余风险或验证盲区。

约束

  • 不要把 docs/plans/*.mddocs/solutions/*.md 当成应删除文件
  • 不要自动进入执行阶段
  • 审查结束后,如果用户要继续实现,提醒其回到 Claude 或继续人工决策
  • 不要把低把握度推断包装成高置信结论

完成后的引导

完成后告诉用户:

  • 主要 findings
  • 是否建议修改 plan / code
  • 如果下一步是执行,应回到 Claude

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.