ghidra-mcp
Ghidra MCP Server — 200+ MCP tools for AI-powered reverse engineering. GUI plugin + headless server, lazy tool loading, convention enforcement, batch operations
What is ghidra-mcp?
Ghidra MCP Server — 200+ MCP tools for AI-powered reverse engineering. GUI plugin + headless server, lazy tool loading, convention enforcement, batch operations, Ghidra Server integration, and Docker deployment.
Overview of ghidra-mcp
What it is
Ghidra MCP Server is a production-ready Model Context Protocol (MCP) server that bridges Ghidra's reverse engineering capabilities with AI tools and automation frameworks. It provides 209 MCP tools covering binary analysis, dynamic analysis, and AI-powered workflows. The project includes a GUI plugin and a headless server, supports lazy tool loading, convention enforcement, batch operations, Ghidra Server integration, and Docker deployment.
Key features
- 209 MCP tools — comprehensive API surface covering function analysis, data flow, structure discovery, string extraction, import/export analysis, memory inspection, and cross-binary documentation
- Dynamic analysis: P-code emulation via EmulatorHelper and live debugger integration with 16 /debugger/* endpoints over Ghidra's TraceRmi framework (GUI plugin only)
- Convention enforcement with three tiers: auto-fix, warn, and reject — naming, typing, and documentation standards are enforced at the tool layer
- Production-grade reliability: atomic transactions, batch operations (93% API call reduction), configurable timeouts, and graceful error handling
- Cross-binary documentation transfer using SHA-256 function hash matching
- Headless and GUI modes, Docker-ready for CI/CD pipelines
Use cases
- AI-powered reverse engineering with step-by-step documentation workflows, Hungarian notation, type auditing, and automated verification scoring
- Batch documentation of multiple functions via parallel subagent dispatch
- Orphaned code discovery in gaps between known code
- Cross-version function matching and documentation propagation
- Multi-user collaboration through Ghidra Server integration with version control and checkout/checkin workflows
- Automated analysis at scale using headless mode and Docker
Getting started
Prerequisites: Java 21 LTS, Apache Maven 3.9+ (or Gradle wrapper), Ghidra 12.1.4, Python 3.10+ with uv or pip + venv.Installation command (recommended for all platforms):python -m tools.setupFor full installation instructions, see the repository README.
License and pricing
Licensed under Apache-2.0. Open source.
Based on the project's public README and repository metadata.