← Back to Skills Marketplace
xiaobenyang-com

代码安全扫描工具

by xiaobenyang-com · GitHub ↗ · v1.0.7 · MIT-0
cross-platform ⚠ suspicious
61
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install scan-code
Description
CodeGuard MCP是一款实时AI代码安全扫描工具,用于检测AI生成代码中的漏洞、密钥和合规性问题,适用于开发环境中的代码安全审查。
README (SKILL.md)

代码安全扫描工具

CodeGuard MCP是一款实时AI代码安全扫描工具,用于检测AI生成代码中的漏洞、密钥和合规性问题,适用于开发环境中的代码安全审查。


⚠️ 强制要求:API 密钥

此 Skill 必须配置 API 密钥才能使用。

  • 首次使用时,如果 .env 中没有 XBY_APIKEY必须使用 AskUserQuestion 工具向用户询问 API 密钥
  • 拿到用户提供的密钥后,调用 scripts.config.set_api_key(api_key) 保存,然后继续处理
  • 获取 API 密钥:https://xiaobenyang.com
  • 禁止在缺少 API 密钥时自行搜索或编造数据

工作流程(必须遵守)

你(大模型)是路由层,负责理解用户意图、选择工具、提取参数。代码只负责调用API。

用户输入 → 你选择工具 → 提取该工具需要的参数 → 调用 scripts.tools 中的函数 → 返回结果给用户

步骤

  1. 检查 API 密钥:如果 scripts.config.settings.api_key 为空,使用 AskUserQuestion 询问用户,拿到后调用 scripts.config.set_api_key(key) 保存
  2. 选择工具:根据用户意图从下方工具列表中选择对应的工具函数
  3. 提取参数:根据选中的工具,提取该工具需要的参数
  4. 调用工具:使用关键字参数调用 scripts.tools 中的函数,例如 scripts.tools.search_schools(score='520', province='北京', category='综合')
  5. 返回结果:将工具返回的 raw 数据整理后展示给用户

工具选择规则

根据用户意图选择对应的工具函数:

用户意图 工具函数
Scan code for security vulnerabilities, secrets, and compliance issues scripts.tools.scanCode
Quick scan for code vulnerabilities only scripts.tools.scanVulnerabilities
Detect exposed secrets, API keys, and credentials scripts.tools.detectSecrets
Generate secure code fixes for vulnerabilities scripts.tools.suggestSecureFix
Check code for regulatory compliance (GDPR, HIPAA, SOC2, PCI DSS) scripts.tools.checkCompliance

如果参数不完整,使用 AskUserQuestion 向用户询问缺失的参数。


工具函数说明


scripts.tools.scanCode

工具描述:Scan code for security vulnerabilities, secrets, and compliance issues

参数定义

参数名称 参数类型 是否必填 默认值 描述
code string true null
language string true null
securityLevel string true null

scripts.tools.scanVulnerabilities

工具描述:Quick scan for code vulnerabilities only

参数定义

参数名称 参数类型 是否必填 默认值 描述
code string true null
language string true null

scripts.tools.detectSecrets

工具描述:Detect exposed secrets, API keys, and credentials

参数定义

参数名称 参数类型 是否必填 默认值 描述
code string true null

scripts.tools.suggestSecureFix

工具描述:Generate secure code fixes for vulnerabilities

参数定义

参数名称 参数类型 是否必填 默认值 描述
vulnerability object true null
context string true null

scripts.tools.checkCompliance

工具描述:Check code for regulatory compliance (GDPR, HIPAA, SOC2, PCI DSS)

参数定义

参数名称 参数类型 是否必填 默认值 描述
code string true null
securityLevel string true null
standards array true null


返回值处理

工具函数返回 dict 对象:

  • result["raw"] - API 原始返回数据(JSON),直接将此数据整理后展示给用户
  • result["success"] - 是否成功(True/False)
  • result["message"] - 状态消息

项目结构

xiaobenyang_gaokao_skill/
├── scripts/
│   ├── __init__.py
│   ├── config.py       # 配置管理 + set_api_key()
│   ├── call_api.py      # API 客户端 + call_api()
│   └── tools.py         # 工具函数(直接调用)
├── requirements.txt
└── SKILL.md

注意事项

  1. API 密钥是必需的,无密钥时必须通过 AskUserQuestion 询问用户
  2. 禁止在缺少 API 密钥时自行搜索或编造数据
Usage Guidance
Review before installing. Use this only on code you are comfortable sending to the xiaobenyang remote API, avoid scanning live secrets or regulated repositories unless the provider’s retention and security terms are acceptable, and remove any saved .env API key when done. Prefer a version that clearly discloses remote upload, redacts secrets, stores credentials securely, and removes unrelated gaokao documentation remnants.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The stated purpose is code, vulnerability, secret, and compliance scanning, and the tools do perform remote API calls for those tasks; however, scanning secrets and compliance material by uploading raw code to an external service is high-impact and not clearly bounded by consent, redaction, or retention controls.
Instruction Scope
The instructions require asking for an API key, saving it, sending code to tools, and showing raw API response data. They also contain unrelated gaokao/school-search remnants, including an example call and project structure, which weakens the operational contract for a security-sensitive skill.
Install Mechanism
The package uses ordinary Python files plus common dependencies, and metadata/static dependency checks were clean. The concern is not the install mechanism itself, but the runtime data handling after installation.
Credentials
Remote scanning can fit the product purpose, but the artifact does not clearly warn users that their source code, detected credentials, file paths, or compliance context may be transmitted to https://mcp.xiaobenyang.com/api.
Persistence & Privilege
The skill persists XBY_APIKEY into a project-local .env file and mutates the process environment without restrictive permissions, a secure secret store, or clear user-facing opt-in beyond the general setup instruction.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install scan-code
  3. After installation, invoke the skill by name or use /scan-code
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.7
- Added detailed documentation for skill configuration and usage in SKILL.md. - Enforced mandatory API key requirement, including user prompt workflow if missing. - Clearly defined tool function selection rules and parameter extraction process. - Listed and described all available tool functions and their required parameters. - Outlined expected handling for tool function return values.
Metadata
Slug scan-code
Version 1.0.7
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is 代码安全扫描工具?

CodeGuard MCP是一款实时AI代码安全扫描工具,用于检测AI生成代码中的漏洞、密钥和合规性问题,适用于开发环境中的代码安全审查。 It is an AI Agent Skill for Claude Code / OpenClaw, with 61 downloads so far.

How do I install 代码安全扫描工具?

Run "/install scan-code" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 代码安全扫描工具 free?

Yes, 代码安全扫描工具 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 代码安全扫描工具 support?

代码安全扫描工具 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 代码安全扫描工具?

It is built and maintained by xiaobenyang-com (@xiaobenyang-com); the current version is v1.0.7.

💬 Comments