The Crash MCP Server is a unified platform for analyzing Linux system crash dumps by integrating the crash utility and drgn programmable debugger through an MCP interface.

Core Capabilities:

• Crash dump analysis: Analyze system crash dumps (vmcore files) using crash utility commands and drgn Python debugging

• Session management: Start, manage, and terminate analysis sessions with automatic deduplication and reuse for the same vmcore

• Remote analysis: Connect to remote hosts via SSH to analyze crash dumps without downloading files locally

• Command execution: Run crash commands (sys, bt, log, ps, files, vm, kmem) and drgn Python code for advanced debugging

• File discovery: Recursively scan directories to locate crash dump files

• Output management: Persist command outputs to disk with configurable truncation, pagination support, and regex search capabilities

• Multi-architecture support: Automatically detect vmcore architecture and compile/use appropriate crash versions (x86_64, ARM64) with support for different compression formats like LZO

• Command caching: Cache outputs for improved performance on repeated queries

• Dual transport modes: Operate in stdio (default) or SSE/HTTP modes for different deployment scenarios

• Configuration flexibility: Customize behavior through environment variables for caching, logging, and output truncation

Provides tools for analyzing Linux system crash dumps through the crash utility, enabling automatic discovery of vmcore files, kernel matching, and execution of crash analysis commands like sys, bt, and ps.

1. Click on "Install Server".

2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.

3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Crash MCP Server analyze the latest crash dump and show the backtrace"

That's it! The server will respond to your query, and you can continue using it as needed.

Here is a step-by-step guide with screenshots.