Happy slice of mcp pizzaMCP.pizza
Fire in da houseTop Tip:Paying $100+ per month for Perplexity, MidJourney, Runway, ChatGPT is crazy - get all your AI tools in one site starting at $15 per month with Galaxy AIFire in da houseCheck it out free

x64dbgMCP

MCP.Pizza Chef: Wasdubya

x64dbgMCP is an MCP server plugin for the x64dbg debugger that exposes key debugging functions via the Model Context Protocol. It currently supports executing debugger commands, checking debugger activity, retrieving module lists, and finding memory base addresses. This enables real-time interaction with x64dbg through LLMs, facilitating dynamic analysis and debugging workflows. The plugin aims to expand support for more x64dbg SDK functions to enhance conversational debugging capabilities.

Use This MCP server To

Execute debugger commands via natural language requests Check if the debugger is currently active Retrieve a list of loaded modules in the debugged process Find base addresses of memory regions dynamically Integrate x64dbg debugging context into LLM workflows Automate dynamic analysis tasks using conversational AI Extend debugging capabilities with LLM-driven queries

README

x64dbgmcp

Model Context Protocol for x64dbg

Working Minimal Version

This plug-in currently only supports four functions from the x64dbgSDK which are:

  1. DbgCmdExec
  2. DbgIsActive
  3. GetModuleList
  4. FindMemBaseAddr

This also only works for x64dbg AFAIK, if you want x86 you may be able to just compile with -D BUILD_X64=OFF

I want to add most of the functions provided by the sdk so dynamic analysis can be as easy as conversing with an LLM. Link to functions here

Most of the HTTP server code was generated by Claude Sonnet 3.7.

This is my first repo, so I realize I may have made some mistakes. Any advice or additional functionality is welcome.

Easy set-up

  1. Grab the .dp64 file from the build\release dir and move it to your defualtx64dbgdir\release\x64\plugins dir
  2. Grab the x64dbgmcp.py file from src dir and map your claude_desktop_config file to its install location
  3. Start x64dbg , then start Claude Desktop. You can always check the logs to make sure the plug-in loaded successfully by pressing ALT+L in x64dbg.

BUILD

  1. clone repo
  2. cd build
  3. cmake .. -DBUILD_X64=ON
  4. cmake --build . --config Release

DEMO

Demo of Plug

x64dbgMCP FAQ

How do I install the x64dbgMCP plugin?
Copy the .dp64 file to your x64dbg plugins directory and configure the Python client with your LLM setup.
Does x64dbgMCP support both x64 and x86 architectures?
Currently, it only supports x64dbg (64-bit). For x86, you may need to compile with the BUILD_X64=OFF flag.
What x64dbg functions does x64dbgMCP support?
It supports DbgCmdExec, DbgIsActive, GetModuleList, and FindMemBaseAddr functions from the x64dbg SDK.
Can I extend x64dbgMCP to support more debugger functions?
Yes, the plugin is designed to be extended with additional x64dbg SDK functions for richer dynamic analysis.
Is x64dbgMCP compatible with multiple LLM providers?
Yes, it can be integrated with various LLMs like OpenAI, Claude, and Gemini through the MCP client.
What is required to start using x64dbgMCP with an LLM?
You need to install the plugin, configure the Python client with your LLM credentials, and start x64dbg and the client.
Does x64dbgMCP provide real-time debugging context to LLMs?
Yes, it exposes live debugging data and commands to LLMs for interactive dynamic analysis.
Is the x64dbgMCP plugin open source?
Yes, it is available on GitHub and welcomes contributions and feedback.