VulnClaw

基于 AI Agent + MCP 工具链 + 渗透 Skill 编排, 配合大语言模型, 自然语言输入 → 自动完成「信息收集 → 漏洞发现 → 漏洞利用 → 报告生成」全流程。

What is VulnClaw?

基于 AI Agent + MCP 工具链 + 渗透 Skill 编排, 配合大语言模型, 自然语言输入 → 自动完成「信息收集 → 漏洞发现 → 漏洞利用 → 报告生成」全流程。

Overview of VulnClaw

What it is

VulnClaw is an AI-driven penetration testing CLI tool. It uses an AI Agent with MCP toolchain and penetration skill orchestration, combined with large language models. Natural language input is automatically processed through the full workflow: information gathering, vulnerability discovery, exploitation, and report generation.

Key features

  • Model-driven solving engine: autonomous loop where the model decides next steps, tool calls, and completion.
  • AgentState evidence memory: tool results are stored in AgentState.evidence with raw text preserved; active context injects high-signal previews only.
  • Natural language driven: describe penetration intent in plain language, automatically identify phases and tools.
  • 13 LLM providers supported: OpenAI, Anthropic, MiniMax, DeepSeek, Zhipu, Moonshot, Qwen, SiliconFlow, Doubao, Stepfun, SenseTime, OpenRouter, local Ollama.
  • MCP toolchain: 4 MCP services (fetch, memory, chrome-devtools, burp) for browser automation and HTTP replay.
  • Vulnerability detection plugin system: low-coupling plugin runtime with built-in read-only web plugins, results automatically integrated into report chain.
  • 50 specialized skills covering CTF, Web, internal network, reverse engineering, vulnerability verification, and authorized red team knowledge base.

Use cases

  • Authorized penetration testing - CTF competitions - Security education and training - Red team exercises

Getting started

bash # Install from PyPI pip install vulnclaw # Or install from source git clone https://github.com/Netw0rkNoob/VulnClaw.git cd VulnClaw pip install -e . # Quick start (four steps) vulnclaw config provider minimax vulnclaw config set llm.api_key sk-your-key-here vulnclaw # Optional: vulnclaw tui

License and pricing

MIT License, open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/netw0rknoobvulnclaw.svg)](https://yunodb.com/tools/netw0rknoobvulnclaw?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/netw0rknoobvulnclaw?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/netw0rknoobvulnclaw.svg" alt="Featured on YunoDB" height="20"></a>