Practical overview
When it fits
Use it when an MCP-compatible AI client needs to work with Ghidra for binary analysis, or when you are setting up an automated Ghidra workflow that benefits from batch operations or headless operation.
Who it is for
- Reverse engineers using Ghidra who want an MCP-compatible AI client to assist with binary analysis and edits.
- Developers building scripted or server-based Ghidra analysis workflows.
- Teams seeking tool-level enforcement of naming and type conventions in AI-assisted changes.
Documented uses
- Expose Ghidra analysis tools to an MCP-compatible AI client.
- Perform batch reverse-engineering operations instead of issuing repeated individual calls.
- Use Ghidra in headless or GUI-plugin mode for interactive or automated analysis.
- Apply naming and type conventions through automatic corrections, warnings or rejected changes.
- Connect workflows to Ghidra Server repositories and collaboration functions.
Requirements & limits
The project is published under the Apache-2.0 license. The documented setup requires a local Ghidra installation and developer tooling.
- An MCP-compatible AI client is needed to use the server as an AI tool bridge.
- The project is tied to Ghidra; the README lists Ghidra 12.1.4 or a compatible version.
- The GUI plugin and headless server support different interactive and automated workflows.
- The README documents Java 21 and Python 3.10 or later; Maven 3.9 or later applies to its Maven-based setup path, while a Gradle wrapper is also provided.
- Shared Ghidra Server users should account for the README's server-version compatibility note.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Build and deploy the Ghidra extension
Prepare the project, build its extension and install it into a local Ghidra installation using the repository setup commands.
Before you start
- Java 21
- Ghidra 12.1.4 or a compatible version
- Python 3.10 or later with uv or pip and a virtual environment
- Maven 3.9 or later for the documented setup-command build path
Steps
Clone the repository and change into its directory.
Run the environment preflight command with the path to the Ghidra installation.
Run the command below.
python -m tools.setup ensure-prereqs --ghidra-pathRun the command below followed by the command below.
python -m tools.setup buildpython -m tools.setup deploy --ghidra-path
Expected outcome
The setup process builds the extension and deploys it to the specified Ghidra installation; the README says deployment starts Ghidra and checks MCP health and schema.
Guide limitations
- Use a compatible Ghidra installation; the README names 12.1.4.
- For shared Ghidra Server use, the README specifies compatible server-version requirements for Ghidra 12.1.4 clients.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 9 October 2026
- Original website title
- bethington/ghidra-mcp · Official homepage title · retained in its source language

