pwndbg-mcp
Enables AI agents to control pwndbg for collaborative debugging with real-time state synchronization and full GDB/pwndbg command support.
README
pwndbg-mcp
MCP Server for pwndbg - 让 AI Agent 能够通过标准 MCP 协议控制 pwndbg 进行协作调试
特性
- ✅ 标准 HTTP MCP 协议
- ✅ 支持远程访问
- ✅ 自动启动,无需手动初始化
- ✅ 完整的 GDB/pwndbg 命令支持
- ✅ 支持人机协作调试
- ✅ 实时状态同步
- ✅ 模块化架构,易于扩展
快速开始
安装
运行安装脚本:
python3 setup.py
按照提示操作,脚本会:
- 检查依赖
- 自动添加到 ~/.gdbinit
- 配置自动加载
手动安装
在 ~/.gdbinit 中添加:
source /path/to/pwndbg-mcp/pwndbg_mcp.py
使用
1. 启动 pwndbg
gdb ./your_binary
HTTP MCP 服务器会自动启动在 127.0.0.1:22334
2. 管理命令
# 查看状态
pwndbg> mcp-status
# 启动服务器(支持自定义配置)
pwndbg> mcp-start # 默认 127.0.0.1:22334
pwndbg> mcp-start 0.0.0.0 # 监听所有接口,端口22334
pwndbg> mcp-start 0.0.0.0 9000 # 自定义host和port
pwndbg> mcp-start 9000 # 只改端口
# 停止服务器
pwndbg> mcp-stop
3. 配置 Claude Code
在 Claude Code 的 MCP 配置中添加:
{
"mcpServers": {
"pwndbg-mcp": {
"url": "http://127.0.0.1:22334"
}
}
}
4. 开始调试
在 Claude Code 中可以直接使用 MCP 工具控制 pwndbg!
支持的工具
| 工具 | 描述 |
|---|---|
gdb_execute |
执行 GDB/pwndbg 命令 |
gdb_get_registers |
获取寄存器状态 |
gdb_get_memory |
读取内存 |
gdb_set_breakpoint |
设置断点 |
gdb_delete_breakpoint |
删除断点 |
gdb_get_breakpoints |
列出所有断点 |
gdb_continue |
继续执行 |
gdb_step |
单步进入 |
gdb_stepi |
单步执行指令 |
gdb_next |
单步跳过 |
gdb_nexti |
下一条指令 |
gdb_get_backtrace |
获取调用栈 |
gdb_get_info |
获取程序信息 |
架构
模块化设计
pwndbg_plugin/
├── __init__.py - 插件入口,自动启动
├── http_bridge.py - HTTP MCP 服务器
├── commands.py - 用户管理命令
├── wrappers.py - 内部包装命令
├── queries.py - 查询接口
├── events.py - 事件处理
└── utils.py - 工具函数
通信协议
- 传输: HTTP POST
- 协议: 标准 MCP (Model Context Protocol)
- 端点:
POST /或POST /mcp - 格式: JSON-RPC 2.0
配置
安全建议
默认监听 127.0.0.1(仅本地访问)。如需远程访问:
pwndbg> mcp-start 0.0.0.0 22334
⚠️ 警告:开放远程访问前请确保网络安全
防火墙配置
如果需要远程访问,确保防火墙允许端口:
# UFW
sudo ufw allow 22334/tcp
# iptables
sudo iptables -A INPUT -p tcp --dport 22334 -j ACCEPT
故障排查
检查服务器状态
pwndbg> mcp-status
测试连接
curl -X POST http://127.0.0.1:22334/ \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"initialize","params":{},"id":1}'
查看日志
服务器输出会显示在 pwndbg 终端中
开发
添加新工具
- 在
queries.py中添加查询函数 - 在
http_bridge.py的_get_tools_list()中注册工具 - 在
_execute_tool()中添加工具映射
测试
cd examples
gdb test_program
pwndbg> start
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.