clawlock

Scan Claw-family deployments for security risks and hardening guidance.

12|Updated Mar 27, 2026
One-click install
npx skills add https://github.com/g1at/ClawLock --skill clawlock
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: clawlock
Source: https://github.com/g1at/ClawLock/tree/main/skill
Command: npx skills add https://github.com/g1at/ClawLock --skill clawlock

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

ClawLock 提供面向 Claw-family 部署的综合安全解决方案,解决跨平台环境中安全扫描、红队演练与系统加固的需求,帮助企业快速发现配置风险、潜在漏洞与合规问题,并给出清晰的改进路径。

Core Features & Use Cases

  • 全量安全扫描与单体技能审计:支持并发执行多域检测,并能对单个 Skill 进行独立审计与导入前检查。
  • 跨平台支持与多种输出:适用于 Linux、macOS、Windows、Android(Termux),输出 JSON、文本与 HTML 报告。
  • 内置深度分析与扩展能力:包含 MCP 深度扫描、OWASP ASI Agent-Scan 和多引擎分析,减少对外部工具的依赖。
  • 离线优先 + 云端扩展:本地离线分析为主,云端威胁情报与 CVE 查询可选加载,灵活适配安全需求。

Quick Start

安装 clawlock 并执行一次全面扫描以评估当前环境的安全性。

Frequently Asked Questions about clawlock

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a comprehensive security scan and hardening check across multiple platforms?

A comprehensive security scan and hardening check requires installing the clawlock package, then executing a full scan to assess your environment's configuration risks, potential vulnerabilities, and compliance issues across Linux, macOS, Windows, and Android.

What is an OWASP ASI Agent-Scan and when do I need it for my deployments?

An OWASP ASI Agent-Scan is a built-in deep analysis feature for red-team readiness that checks agent security. You need it to proactively identify potential vulnerabilities and configuration risks in your Claw-family deployments without relying on external tools.

Does the security scanner work offline or do I need a cloud connection for CVE queries?

The security scanner works offline by default for local analysis, but offers optional cloud intelligence for threat intelligence and CVE queries. This flexible approach adapts to your security requirements whether you have network access or not.

Can I audit a single Skill before importing it into my Claw environment?

Yes, you can audit a single Skill before importing it into your Claw environment. The tool supports independent auditing and pre-import checks for individual Skills to ensure they meet your security and compliance standards.

Do I need Python and promptfoo to perform red-team readiness checks?

You need Python >= 3.9, the clawlock package, and the clawlock binary for basic red-team readiness checks. The promptfoo dependency is optional and only required if you want to enable advanced red-team workflows.

What is the best way to get JSON, text, and HTML reports for MCP deep scanning?

The best way to get JSON, text, and HTML reports for MCP deep scanning is to run the built-in multi-engine analysis. It generates these multiple output formats natively, reducing your dependency on external reporting tools.