by svnscha
Model Context Protocol for WinDBG
# Add to your Claude Code skills
git clone https://github.com/svnscha/mcp-windbgA Model Context Protocol server that bridges AI models with WinDbg for crash dump analysis and remote debugging.
This MCP server integrates with CDB to enable AI models to analyze Windows crash dumps and connect to remote debugging sessions using WinDbg/CDB.
An AI-powered tool that bridges LLMs with WinDbg for crash dump analysis and live debugging. Execute debugger commands through natural language queries like "Show me the call stack and explain this access violation".
Not a magical auto-fix solution. It's a Python wrapper around CDB that leverages LLM knowledge to assist with debugging.
[!TIP] In enterprise environments, MCP server usage might be restricted by organizational policies. Check with your IT team about AI tool usage and ensure you have the necessary permissions before proceeding.
No comments yet. Be the first to share your thoughts!
pip install mcp-windbg
The MCP server supports multiple transport protocols:
| Transport | Description | Use Case |
|-----------|-------------|----------|
| stdio (default) | Standard input/output | Local MCP clients like VS Code, Claude Desktop |
| streamable-http | Streamable HTTP | Modern HTTP clients with bidirectional streaming |
Standard I/O (default):
mcp-windbg
# or explicitly
mcp-windbg --transport stdio
Streamable HTTP:
mcp-windbg --transport streamable-http --host 127.0.0.1 --port 8000
Endpoint: http://127.0.0.1:8000/mcp
--transport {stdio,streamable-http} Transport protocol (default: stdio)
--host HOST HTTP server host (default: 127.0.0.1)
--port PORT HTTP server port (default: 8000)
--cdb-path PATH Custom path to cdb.exe
--symbols-path PATH Custom symbols path
--timeout SECONDS Command timeout (default: 30)
--verbose Enable verbose output
To make MCP servers available in all your workspaces, use the global user configuration:
F1, type > and select MCP: Open User Configuration.{
"servers": {
"mcp_windbg": {
"type": "stdio",
"command": "python",
"args": ["-m", "mcp_windbg"],
"env": {
"_NT_SYMBOL_PATH": "SRV*C:\\Symbols*https://msdl.microsoft.com/download/symbols"
}
}
}
}
This enables MCP Windbg in any workspace, without needing a local .vscode/mcp.json file.
For scenarios where you need to run the MCP server separately (e.g., remote access, shared server, or debugging the server itself), you can use the HTTP transport:
1. Start the server manually:
python -m mcp_windbg --transport streamable-http --host 127.0.0.1 --port 8000
2. Configure VS Code to connect via HTTP:
{
"servers": {
"mcp_windbg_http": {
"type": "http",
"url": "http://localhost:8000/mcp"
}
}
}
Workspace-specific and alternative configuration: See Installation documentation for details on configuring Claude Desktop, Cline, and other clients, or for workspace-only setup.
Once configured, restart your MCP client and start debugging:
Analyze the crash dump at C:\dumps\app.dmp
This server implements the Model Context Protocol (MCP), making it compatible with any MCP-enabled client:
The beauty of MCP is that you write the server once, and it works everywhere. Choose your favorite AI assistant!
| Tool | Purpose | Use Case |
|------|---------|----------|
| list_windbg_dumps | List crash dump files | Discovery and batch analysis |
| open_windbg_dump | Analyze crash dumps | Initial crash dump analysis |
| close_windbg_dump | Cleanup dump sessions | Resource management |
| open_windbg_remote | Connect to remote debugging | Live debugging sessions |
| close_windbg_remote | Cleanup remote sessions | Resource management |
| run_windbg_cmd | Execute WinDbg commands | Custom analysis and investigation |
| send_ctrl_break | Break into a running target | Interrupt execution during live debugging |
| Topic | Description | |-------|-------------| | Getting Started | Quick setup and first steps | | Installation | Detailed installation for pip, MCP registry, and from source | | Usage | MCP client integration, command-line usage, and workflows | | Tools Reference | Complete API reference and examples | | Troubleshooting | Common issues and solutions |
Analyze this heap address with !heap -p -a 0xABCD1234 and check for buffer overflow"
Execute !peb and tell me if there are any environment variables that might affect this crash"
Run .ecxr followed by k and explain the exception's root cause"
"Connect to tcp:Port=5005,Server=192.168.0.100 and show me the current thread state"
"Send CTRL+BREAK to the live session, then dump all thread stacks with ~*k"
"Check for timing issues in the thread pool with !runaway and !threads"
"Show me all threads with ~*k and identify which one is causing the hang"
Read about the development journey: The Future of Crash Analysis: AI Meets WinDbg
MIT