skill-vetter

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Skill Vetter

Skill 审核工具

Security-first vetting protocol for AI agent skills. Never install a skill without vetting it first.

以安全为核心的AI Agent技能审核规范。未经审核绝不要安装任何skill。

When to Use

使用场景

Before installing any skill from community marketplaces
Before running skills from GitHub repos
When evaluating skills shared by other agents
Anytime you're asked to install unknown code

从社区市场安装任何skill之前
运行GitHub仓库中的skill之前
评估其他Agent分享的skill时
任何被要求安装未知代码的场景

Vetting Protocol

审核流程

Step 1: Source Check

步骤1：来源核查

Questions to answer:
- [ ] Where did this skill come from?
- [ ] Is the author known/reputable?
- [ ] How many downloads/stars does it have?
- [ ] When was it last updated?
- [ ] Are there reviews from other agents?

需要回答的问题：
- [ ] 该skill来自何处？
- [ ] 作者是否知名/可信？
- [ ] 它的下载量/星标数有多少？
- [ ] 最后更新时间是什么时候？
- [ ] 是否有其他Agent的评价？

Step 2: Code Review (MANDATORY)

步骤2：代码审查（强制要求）

Read ALL files in the skill. Check for these RED FLAGS:

REJECT IMMEDIATELY IF YOU SEE:
─────────────────────────────────────────
• curl/wget to unknown URLs
• Sends data to external servers
• Requests credentials/tokens/API keys
• Reads ~/.ssh, ~/.aws, ~/.config without clear reason
• Accesses MEMORY.md, USER.md, SOUL.md, IDENTITY.md
• Uses base64 decode on anything
• Uses eval() or exec() with external input
• Modifies system files outside workspace
• Installs packages without listing them
• Network calls to IPs instead of domains
• Obfuscated code (compressed, encoded, minified)
• Requests elevated/sudo permissions
• Accesses browser cookies/sessions
• Touches credential files
─────────────────────────────────────────

阅读skill中的所有文件，检查以下危险信号：

出现以下情况立即拒绝：
─────────────────────────────────────────
• 使用curl/wget访问未知URL
• 向外部服务器发送数据
• 请求凭证/令牌/API密钥
• 无明确理由读取~/.ssh、~/.aws、~/.config
• 访问MEMORY.md、USER.md、SOUL.md、IDENTITY.md
• 对任何内容使用base64解码
• 对外部输入使用eval()或exec()
• 修改工作区外的系统文件
• 未列出依赖就安装包
• 通过IP而非域名发起网络请求
• 混淆代码（压缩、编码、混淆）
• 请求提升权限/sudo权限
• 访问浏览器Cookie/会话
• 触碰凭证文件
─────────────────────────────────────────

Step 3: Permission Scope

步骤3：权限范围评估

Evaluate:
- [ ] What files does it need to read?
- [ ] What files does it need to write?
- [ ] What commands does it run?
- [ ] Does it need network access? To where?
- [ ] Is the scope minimal for its stated purpose?

评估内容：
- [ ] 它需要读取哪些文件？
- [ ] 它需要写入哪些文件？
- [ ] 它会运行哪些命令？
- [ ] 是否需要网络访问？访问哪里？
- [ ] 权限范围是否与其宣称的用途最小化？

Step 4: Risk Classification

步骤4：风险等级划分

Risk Level	Examples	Action
LOW	Notes, weather, formatting	Basic review, install OK
MEDIUM	File ops, browser, APIs	Full code review required
HIGH	Credentials, trading, system	Human approval required
EXTREME	Security configs, root access	Do NOT install

风险等级	示例	操作建议
低	笔记、天气、格式处理	基础审核，可安装
中	文件操作、浏览器、API调用	需完整代码审查
高	凭证处理、交易、系统操作	需人工批准
极高	安全配置、 root权限	禁止安装

Output Format

输出格式

After vetting, produce this report:

SKILL VETTING REPORT
═══════════════════════════════════════
Skill: [name]
Source: [community / GitHub / other]
Author: [username]
Version: [version]
───────────────────────────────────────
METRICS:
• Downloads/Stars: [count]
• Last Updated: [date]
• Files Reviewed: [count]
───────────────────────────────────────
RED FLAGS: [None / List them]

PERMISSIONS NEEDED:
• Files: [list or "None"]
• Network: [list or "None"]  
• Commands: [list or "None"]
───────────────────────────────────────
RISK LEVEL: [LOW / MEDIUM / HIGH / EXTREME]

VERDICT: [SAFE TO INSTALL / INSTALL WITH CAUTION / DO NOT INSTALL]

NOTES: [Any observations]
═══════════════════════════════════════

审核完成后，生成如下报告：

SKILL 审核报告
═══════════════════════════════════════
Skill: [名称]
来源: [社区 / GitHub / 其他]
作者: [用户名]
版本: [版本号]
───────────────────────────────────────
指标：
• 下载量/星标数: [数量]
• 最后更新时间: [日期]
• 已审查文件数: [数量]
───────────────────────────────────────
危险信号: [无 / 列出具体内容]

所需权限：
• 文件: [列表或“无”]
• 网络: [列表或“无”]  
• 命令: [列表或“无”]
───────────────────────────────────────
风险等级: [低 / 中 / 高 / 极高]

审核结论: [可安全安装 / 谨慎安装 / 禁止安装]

备注: [任何观察结果]
═══════════════════════════════════════

Quick Vet Commands

快速审核命令

For GitHub-hosted skills:

bash

undefined

针对GitHub托管的skill：

bash

undefined

Check repo stats

检查仓库统计信息

curl -s "https://api.github.com/repos/OWNER/REPO" | jq '{stars: .stargazers_count, forks: .forks_count, updated: .updated_at}'

List skill files

列出skill文件

curl -s "https://api.github.com/repos/OWNER/REPO/contents/skills/SKILL_NAME" | jq '.[].name'

Fetch and review SKILL.md

获取并查看SKILL.md

curl -s "https://raw.githubusercontent.com/OWNER/REPO/main/skills/SKILL_NAME/SKILL.md"

undefined

curl -s "https://raw.githubusercontent.com/OWNER/REPO/main/skills/SKILL_NAME/SKILL.md"

undefined

Trust Hierarchy

信任层级

Official marketplace skills - Lower scrutiny (still review)
High-star repos (1000+) - Moderate scrutiny
Known authors - Moderate scrutiny
New/unknown sources - Maximum scrutiny
Skills requesting credentials - Human approval always

官方市场skill - 较低审查强度（仍需审核）
高星仓库（1000+星） - 中等审查强度
知名作者 - 中等审查强度
新/未知来源 - 最高审查强度
请求凭证的skill - 始终需要人工批准

Remember

注意事项

No skill is worth compromising security
When in doubt, don't install
Ask your human for high-risk decisions
Document what you vet for future reference

Paranoia is a feature.

没有任何skill值得牺牲安全
存疑时，不要安装
高风险决策请咨询人类用户
记录审核内容以备未来参考