13bm/GhidraMCP

Overview

GhidraMCP is an open-source Java server that implements the Model Context Protocol over sockets for the Ghidra reverse engineering framework. It enables AI assistants to interact with Ghidra programmatically, allowing automated analysis and script execution.

Best for

Best for
Reverse engineers and security researchers using Ghidra who want to augment their workflow with AI agents.

Use cases

• Integrating AI coding assistants with Ghidra for automated reverse engineering
• Triggering Ghidra scripts or analyses from an AI agent via MCP calls
• Building custom tooling that combines Ghidra's capabilities with large language models

How to use

Tools exposed

• list_functions
• list_classes
• list_imports
• list_exports
• list_namespaces
• list_data_items
• list_strings
• search_functions_by_name
• get_function_by_address
• get_current_address
• get_current_function
• decompile_function
• decompile_function_by_address
• disassemble_function
• get_xrefs_to
• get_xrefs_from
• get_function_xrefs
• get_program_info
• get_memory_map
• get_variables

Tested with

Claude Desktop, Claude Code, Cursor, Continue

Example client config

{\n  "mcpServers": {\n    "ghidra": {\n      "command": "/path/to/mcp_bridge",\n      "args": ["--host", "localhost", "--port", "8765"]\n    }\n  }\n}

Notes

GhidraMCP is an open-source Java server that implements the Model Context Protocol over sockets for the Ghidra reverse engineering framework. It enables AI assistants to interact with Ghidra programmatically, allowing automated analysis and script execution.

119 stars on GitHub. Last updated 2026-05-14. Licensed Apache-2.0.

Use cases

• Integrating AI coding assistants with Ghidra for automated reverse engineering
• Triggering Ghidra scripts or analyses from an AI agent via MCP calls
• Building custom tooling that combines Ghidra’s capabilities with large language models

Pros

• Socket-based communication is efficient and language-agnostic for clients
• Active project with 119 stars and ongoing development
• Leverages the MCP standard for interoperability with many AI tools

Cons

• Requires a running Ghidra instance and Java setup, increasing deployment complexity
• Documentation and community support may still be limited compared to larger projects
• Scope is narrow – only useful if you already work with Ghidra and want AI integration

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.