x64dbgMCP
Model Context Protocol for x64dbg & x32dbg
claude mcp add --transport stdio wasdubya-x64dbgmcp python Path\\to\\x64dbgmcp.py \ --env PYTHONUNBUFFERED="1"
How to use
This MCP server bridges the x64dbg debugger with language model prompts to enable automated debugging workflows. It exposes the x64dbgMCP tooling through an MCP interface, allowing you to drive debugging tasks, symbol parsing, thread management, and cross-architecture operations from your preferred prompt-based workflow. The server is designed to work with Claude Desktop and Cursor in the example setup, but can be used by any prompt-driven environment that can consume MCP endpoints. Install the necessary Python dependencies and run the MCP server to start sending commands that control x64dbg via the MCP bridge.
Once running, you can issue prompts that map to x64dbg SDK tools, such as retrieving register values, performing pattern searches, listing modules, and initiating symbol-based navigation. The server supports multi-arch debugging (x64/x32) and provides features like symbol parsing with pagination and Xrefs by address or string, enabling both quick inspections and deep static/dynamic analysis through natural language prompts.
How to install
Prerequisites:
- Python 3.8+ installed on the host machine
- Access to x64dbg and your debugging targets
Step-by-step installation:
-
Install Python dependencies
- Open a terminal and run: pip install mcp requests
-
Place MCP server script
- Ensure you have the MCP server script (e.g., x64dbgmcp.py) in a known location.
-
Configure the MCP entry point
- Update Path to the Python interpreter and the MCP script in your environment or editor configuration.
- Example path (adjust to your environment): Path\to\python\python.exe and Path\to\x64dbgmcp.py
-
Run the MCP server
- From terminal: python Path\to\x64dbgmcp.py
-
Verify operation
- Connect your MCP client or Claude Desktop instance and ensure the server name (x64dbg) is reachable.
Additional notes
Environment and configuration tips:
- Ensure Python prints are unbuffered to get real-time feedback: export PYTHONUNBUFFERED=1 (or set in your environment).
- If you encounter connection issues, verify that the path to x64dbgmcp.py is correct and that Python can access the x64dbg environment.
- The server is designed to operate with x64dbg across both x64 and x32 processes; ensure your target binaries are loaded appropriately in x64dbg when testing.
- If you need to customize the command, you can point to a virtual environment's Python interpreter or wrap the call in a shell script that sets up the environment before launching the MCP script.
- For troubleshooting, check logs in the MCP client for any errors parsing responses from the Python server or failures to initialize x64dbg features.
Related MCP Servers
xiaozhi-esp32
An MCP-based chatbot | 一个基于MCP的聊天机器人
unreal
Enable AI assistant clients like Cursor, Windsurf and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP).
deeppowers
DEEPPOWERS is a Fully Homomorphic Encryption (FHE) framework built for MCP (Model Context Protocol), aiming to provide end-to-end privacy protection and high-efficiency computation for the upstream and downstream ecosystem of the MCP protocol.
cpp
Lightweight C++ MCP (Model Context Protocol) SDK
mcp-cpp
A C++ SDK for the Model Context Protocol (MCP). The SDK will provide a framework for creating MCP servers and clients in C++.
processhacker
your ai debugger, vibe hacking tool