Ghidra
Enables large language models to autonomously reverse engineer applications by exposing Ghidra functionalities via a Model Context Protocol server.
概要
GhidraMCP acts as a Model Context Protocol (MCP) server, seamlessly integrating with the powerful Ghidra reverse engineering framework. Its core purpose is to empower Large Language Models (LLMs) to autonomously analyze and understand compiled applications. By exposing a comprehensive set of Ghidra's core functionalities—including decompilation, binary analysis, and the listing of methods, classes, imports, and exports—GhidraMCP allows LLM clients to programmatically interact with and extract deep insights from binaries. This integration facilitates advanced AI-driven tasks in reverse engineering, security analysis, and automated software understanding.
主な機能
- Automatically renames methods and data
- Lists methods, classes, imports, and exports
- 2 GitHub stars
- Provides an MCP Server and Ghidra Plugin
- Decompiles and analyzes binaries within Ghidra
ユースケース
- Autonomous reverse engineering by LLMs
- Integrating Ghidra's analysis capabilities with MCP-compatible AI clients
- Automating security analysis and vulnerability research using AI
