← 返回 Skills 市场
Code Review Pro
作者
Softboypatrick
· GitHub ↗
· v1.0.0
· MIT-0
35
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install code-review-pro
功能描述
专业代码审查 | 4Agent并行审查+ 置信度评分 | 覆盖逻辑/安全/性能/风格
使用说明 (SKILL.md)
Code Review Pro
4Agent 并行代码审查。置信度≥70% 才输出。
Agent 1: 逻辑
检查：逻辑错误、边界条件、竞态条件、死循环
Agent 2: 安全
检查：注入漏洞、XSS、CSRF、权限泄露、密钥硬编码
Agent 3: 性能
检查：时间复杂度、内存泄漏、不必要重复、过大循环
Agent 4: 风格
检查：命名规范、代码整洁度、语言惯用法
输出格式
[严重] 问题描述 (置信度: 92%)
- 位置：文件:行号 | 风险：高危 | 建议：...
[建议] 优化点 (置信度: 78%)
- ...\r
安全使用建议
This skill is instruction-only and internally consistent with a multi-aspect code reviewer; it doesn't request credentials or install anything, so risk is low. Still, verify outputs before applying suggested fixes (LLMs can hallucinate or mis-evaluate complex code), avoid sharing sensitive secrets or proprietary code until you're comfortable with results, and prefer testing on non-sensitive samples. Also note the source/author is unknown — consider running a few known test cases to confirm the review quality before using it on important projects.
功能分析
Type: OpenClaw Skill
Name: code-review-pro
Version: 1.0.0
The skill bundle 'Code Review Pro' provides instructions for an AI agent to perform multi-perspective code reviews (logic, security, performance, and style). The content in SKILL.md and _meta.json is purely descriptive, lacks any executable code or suspicious commands, and aligns entirely with its stated purpose without any indicators of malicious intent or risk.
能力评估
Purpose & Capability
Name/description claim a multi-agent code review covering logic, security, performance, and style; the skill is instruction-only and requires no binaries, env vars, or installs — all consistent with a reviewer helper.
Instruction Scope
SKILL.md enumerates four focused review agents and an output format. It does not instruct reading unrelated system files, exfiltrating data, calling external endpoints, or accessing credentials. The checks listed (e.g., injection, XSS, hardcoded keys) are reasonable for a security review.
Install Mechanism
No install spec and no code files — the lowest-risk form. Nothing will be written to disk or fetched at install time.
Credentials
The skill declares no required environment variables, credentials, or config paths. There are no disproportionate secret requests compared to its stated functionality.
Persistence & Privilege
always is false and model invocation is not disabled (the platform default). The skill does not request persistent presence or modify other skills or system settings.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install code-review-pro - 安装完成后，直接呼叫该 Skill 的名称或使用
/code-review-pro触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release of Code Review Pro
- Parallel code review with four specialized agents: Logic, Security, Performance, and Style.
- Outputs issues only with confidence ≥70%.
- Covers logical errors, security vulnerabilities, performance bottlenecks, and code style problems.
- Provides structured feedback with severity, confidence score, location, risk level, and suggestions.
- Supports both severe issues and optimization tips.
元数据
常见问题
Code Review Pro 是什么？
专业代码审查 | 4Agent并行审查+ 置信度评分 | 覆盖逻辑/安全/性能/风格. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 35 次。
如何安装 Code Review Pro？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install code-review-pro」即可一键安装，无需额外配置。
Code Review Pro 是免费的吗？
是的，Code Review Pro 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Code Review Pro 支持哪些平台？
Code Review Pro 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Code Review Pro？
由 Softboypatrick（@softboypatrick）开发并维护，当前版本 v1.0.0。
推荐 Skills