Reverse-engineer binaries in Ghidra

Tools

custom-built or unnamed agent

Overview

If you find this useful, please ⭐ star the repo — it helps others discover it!

Workflow

If you find this useful, please ⭐ star the repo — it helps others discover it! If Ghidra MCP saves you time, consider sponsoring the project. One-time and recurring support both help fund compatibility updates, production hardening, docs, and new tooling. A production-ready Model Context Protocol (MCP) server that bridges Ghidra's powerful reverse engineering capabilities with modern AI tools and automation frameworks. 219 MCP tools, battle-tested AI workflows, and the most comprehensive Ghidra-MCP integration available — now including P-code emulation, live debugger integration, and PCode-graph data flow analysis. Quickstart: install a release (no clone, no build) Every release ships the two pieces you need: the Ghidra extension (GhidraMCP-<version>.zip) and the MCP bridge (ghidra_mcp_bridge-<version>-py3-none-any.whl). Install the prerequisites: Ghidra 12.1.4, Java 21 and uv. The extension only loads in the exact Ghidra version it was built for. Install the prerequisites: Ghidra 12.1.4, Java 21 and uv. The extension only loads in the exact Ghidra version it was built for. Install the extension: in Ghidra's project window, File > Install Extensions, click +, pick the GhidraMCP-<version>.zip you downloaded, and restart Ghidra. Install the extension: in Ghidra's project window, File > Install Extensions, click +, pick the GhidraMCP-<version>.zip you downloaded, and restart Ghidra. Enable the plugin: in the project window, File > Configure > Utility > Configure, tick GhidraMCPPlugin. Check it answers: curl http://127.0.0.1:8089/check_connection # {"status": "ok", "server_kind": "gui", "version": "7.0.0"} Enable the plugin: in the project window, File > Configure > Utility > Configure, tick GhidraMCPPlugin. Check it answers: Install the bridge (copy the wheel's link from the same release page): uv tool install https://github.com/bethington/ghidra-mcp/releases/download/v7.0.0/ghidra_mcp_bridge-7.0.0-py3-none-any.whl Install the bridge (copy the wheel's link from the same release page): Point your MCP client at it, using the absolute path (which bridge-mcp-ghidra, or where.exe bridge-mcp-ghidra on Windows): { "mcpServers": { "ghidra-mcp": { "command": "/home/<you>/.local/bin/bridge-mcp-ghidra" } } } On Windows: "command": "C:\\Users\\<you>\\.local\\bin\\bridge-mcp-ghidra.exe". A client launched from a GUI or a service does not see your shell's PATH, so a bare bridge-mcp-ghidra fails with spawn ENOENT. Point your MCP client at it, using the absolute path (which bridge-mcp-ghidra, or where.exe bridge-mcp-ghidra on Windows): On Windows: "command": "C:\\Users\\<you>\\.local\\bin\\bridge-mcp-ghidra.exe". A client launched from a GUI or a service does not see your shell's PATH, so a bare bridge-mcp-ghidra fails with spawn ENOENT. Open a program in Ghidra and ask your AI tool to list its functions. Screenshots of steps 2-3: docs/INSTALL_GUI.md. Building from source, Docker, the HTTP transports and every bridge flag are further down. Which version should I run? Tagged releases (7.0.0, 7.1.0, ...) are stable. Release candidates and main are beta: feature-complete and live-tested, but shapes can still change before final. dev is experimental: tools are merged, renamed and refactored there freely, so build nothing on its API. Why Ghidra MCP? Most Ghidra MCP implementations give you a handful of read-only tools and call it a day. This project is different — it was built by a reverse engineer who uses it daily on real binaries, not as a demo. 219 MCP tools — 3x more than any competing implementation. Not just read operations — full write access for renaming, typing, commenting, structure creation, script execution, P-code emulation, and live debugging. Battle-tested AI workflows — Proven documentation workflows (V5) refined across hundreds of functions. Includes step-by-step prompts, Hungarian notation reference, batch processing guides, and orphaned code discovery. Production-grade reliability — Atomic transactions, batch operations (93% API call reduction), configurable timeouts, and graceful error handling. No silent failures. Cross-binary documentation transfer — SHA-256 function hash matching propagates documentation across binary versions automatically. Document once, apply everywhere. Full Ghidra Server integration — Connect to shared Ghidra servers, manage repositories, version control, checkout/checkin workflows, and multi-user collaboration.

Reported result

The source claims production-grade reliability with atomic transactions and 93% fewer API calls via batch operations. It states that analyze_function_completeness provides honest 0-100% scoring by forgiving structural deductions. No independent verification of these metrics or actual business impact is provided in the text. A measurable acceptance check would be to compare API call counts before and after enabling batch operations on a sample binary.

Human role

The original post does not say how a person stayed involved.

Limitations

Prerequisites include exact version matching: the extension only loads in the specific Ghidra version it was built for. GUI-launched clients often fail with 'spawn ENOENT' if the bridge path is not absolute. Script execution is opt-in via environment variables. Source reliability is secondary; claims of 'battle-tested' workflows are self-reported by the project author without external audit.

Try it

I need to set up Ghidra MCP for reverse engineering a binary. First, check if I have Ghidra 12.2, Java 25, and uv installed. If not, guide me through installation. Then, download the GhidraMCP extension zip and the bridge wheel from the latest release. Install the extension in Ghidra via File > Install Extensions and restart. Enable the plugin in Configure > Utility. Verify the server is running by curling http://127.0.0.1:8089/check_connection. Next, install the bridge using uv tool install [replaceable wheel URL]. Configure my AI client (e.g., Cursor or Claude) to use the bridge with an absolute path to the executable to avoid PATH issues. Open [replaceable binary file] in Ghidra. Ask the AI to list functions and perform a batch rename operation on the first 5 functions. Before executing any write operations, show me the proposed changes for my approval. If successful, report the number of API calls saved compared to individual operations. If errors occur, check for version mismatches or PATH issues.

More cases · Security & operations

• Run bioinformatics prediction challenges end to end with agents
• Reverse engineer Claude Desktop to find a path to Linux support
• Control and calibrate a robotic hand with OpenClaw
• Figma Security Alert Triage
• Patch laptop BIOS to unlock hidden firmware menus