pyghidra-lite is a token-efficient MCP server designed for Ghidra-based reverse engineering, capable of analyzing ELF, Mach-O, and PE binaries with support for Swift, Objective-C, and Hermes.
From the registry: Token-efficient Ghidra RE: decompile, xrefs, Swift/ObjC, ELF/Mach-O, async analysis
$ brew install openjdk@21$ brew install --cask ghidrahttps://ghidra-sre.org$ pip install pyghidra-litePlease install the `pyghidra-lite` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **JDK 21+** — Install Java 21 or later Run: `brew install openjdk@21`
- **Ghidra 11.x** — Install Ghidra reverse engineering framework Run: `brew install --cask ghidra` (https://ghidra-sre.org)
- **Install pyghidra-lite** — Install from PyPI Run: `pip install pyghidra-lite`
Canonical MCP server config (stdio transport):
- command: `pyghidra-lite`
- args: []
- optional environment variables:
- `GHIDRA_INSTALL_DIR`: Path to Ghidra install dir (auto-detected at /opt/ghidra or ~/ghidra) (example: `/path/to/ghidra`)
Note: Token-efficient Ghidra reverse engineering MCP. Shared JVM proxy architecture. 8 tools. Supports ELF, Mach-O, PE with Swift/ObjC/Hermes detection. Also available via uvx.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.GHIDRA_INSTALL_DIRPath to Ghidra installation if not in standard locationPYGHIDRA_LITE_NO_AUTOSTARTDisable auto-start of the backendAI-powered threat hunting and incident response MCP server for Elasticsearch/OpenSearch