Lisa Py
LLDB MCP Integration + other helpful commands
What is Lisa Py?
Lisa Py is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to lldb mcp integration + other helpful commands
LLDB MCP Integration + other helpful commands
This server falls under the Security and Developer Tools categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- LLDB MCP Integration + other helpful commands
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx lisa-pyConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Lisa Py
Lisa.py is an LLDB MCP integration that bridges the LLDB debugger and AI assistants like Claude, enabling natural language control of debugging sessions through a standardized Model Context Protocol interface. It consists of two components: an MCP server (server.py) that communicates with Claude, and an LLDB plugin (lldb_plugin.py) that runs inside the debugger and exposes over 17 debugger operations via JSON-RPC. Reverse engineers, security researchers, and developers working on native code — especially on Apple Silicon macOS — use it to let Claude set breakpoints, read memory, inspect variables, evaluate expressions, and analyze assembly code alongside them in the debugger.
Prerequisites
- macOS with LLDB installed (comes with Xcode Command Line Tools: `xcode-select --install`)
- Python 3.10 or later with LLDB Python bindings available
- fastmcp Python package version 1.2.0 or later
- httpx Python package
- An MCP client such as Claude Desktop
Clone the lisa.py repository
Clone the repository to a permanent location. The MCP config will reference this path, so choose a stable directory.
git clone https://github.com/ant4g0nist/lisa.py.git
cd lisa.pyInstall Python dependencies
Install the required packages. Use pip or uv, ensuring you install into the Python environment that LLDB's Python bindings use.
pip install "fastmcp>=1.2.0" httpxConfigure Claude Desktop to run the MCP server
Edit ~/Library/Application Support/Claude/claude_desktop_config.json and add the LLDB MCP server entry. Use absolute paths for both uv/python and the repository directory.
{
"mcpServers": {
"lldb": {
"command": "/path/to/your/.local/bin/uv",
"args": [
"--directory",
"/path/to/your/lisa.py/llmcp",
"run",
"lldb_mcp.py"
]
}
}
}Restart Claude Desktop
Fully quit and reopen Claude Desktop to load the new MCP server configuration. You should see the LLDB tools appear (look for the hammer icon).
Load the LLDB plugin in your debug session
Start LLDB with your target binary, then import the plugin script. This starts the JSON-RPC listener that the MCP server communicates with on port 13338.
lldb /path/to/your/binary
(lldb) command script import /path/to/lisa.py/lisa.pyOptional: auto-load the plugin via .lldbinit
To load the plugin automatically every time LLDB starts, add the import command to your ~/.lldbinit file.
echo 'command script import /path/to/lisa.py/lisa.py' >> ~/.lldbinitLisa Py Examples
Client configuration
claude_desktop_config.json entry for the lisa.py LLDB MCP server. Replace paths with your actual uv binary location and repository clone path.
{
"mcpServers": {
"lldb": {
"command": "/Users/yourname/.local/bin/uv",
"args": [
"--directory",
"/Users/yourname/tools/lisa.py/llmcp",
"run",
"lldb_mcp.py"
]
}
}
}Prompts to try
Natural language debugging commands that map to the LLDB MCP tools once the plugin is loaded in your debug session.
- "Debug the program at /tmp/target and set a breakpoint at main"
- "Run the program with arguments -v input.txt"
- "Show me the variables in the current stack frame"
- "Get the backtrace for the current thread"
- "Step into the next function call"
- "Read memory at address 0x100003f00 and show me 32 bytes"
- "Evaluate the expression ptr->data[i] in the current context"
- "Show me the assembly code at the current instruction pointer"Troubleshooting Lisa Py
Claude does not see LLDB tools after configuring the MCP server
Verify the paths in claude_desktop_config.json are absolute and correct. Test by running the command manually: `/path/to/uv --directory /path/to/lisa.py/llmcp run lldb_mcp.py`. If it errors, fix the dependency installation first.
The LLDB plugin fails to connect to the MCP server (port 13338 in use)
Another process is using port 13338. Find and stop it with `lsof -i :13338`. If you need to change the port, look for the port configuration in lldb_plugin.py and update both the plugin and the MCP server connection settings consistently.
command script import fails with 'No module named fastmcp'
The LLDB Python environment is separate from your regular Python installation. Install fastmcp into the LLDB-linked Python: find it with `lldb --python-path` and use that interpreter directly: `/path/to/lldb-python -m pip install fastmcp httpx`.
Frequently Asked Questions about Lisa Py
What is Lisa Py?
Lisa Py is a Model Context Protocol (MCP) server that lldb mcp integration + other helpful commands It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Lisa Py?
Follow the installation instructions on the Lisa Py GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Lisa Py?
Lisa Py works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Lisa Py free to use?
Yes, Lisa Py is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Lisa Py Alternatives — Similar Security Servers
Looking for alternatives to Lisa Py? Here are other popular security servers you can use with Claude, Cursor, and VS Code.
Casdoor
★ 13.6kAn open-source Agent-first Identity and Access Management (IAM) /LLM MCP & agent gateway and auth server with web UI supporting OpenClaw, MCP, OAuth, OIDC, SAML, CAS, LDAP, SCIM, WebAuthn, TOTP, MFA, Face ID, Google Workspace, Azure AD
ghidraMCP
★ 9.0kAn Model Context Protocol server that enables LLMs to autonomously reverse engineer applications by exposing Ghidra's decompilation and analysis tools. It allows AI agents to list code structures, rename methods, and analyze binaries directly through
HexStrike AI
★ 8.9kHexStrike AI MCP Agents is an advanced MCP server that lets AI agents (Claude, GPT, Copilot, etc.) autonomously run 150+ cybersecurity tools for automated pentesting, vulnerability discovery, bug bounty automation, and security research. Seamlessly b
IDA Pro MCP
★ 8.7kEnables AI-assisted reverse engineering in IDA Pro by providing tools to analyze binaries, decompile functions, manage comments, search patterns, and interact with the IDA database through natural language.
Anthropic Cybersecurity Skills
★ 6.6k754 structured cybersecurity skills for AI agents · Mapped to 5 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND & NIST AI RMF · agentskills.io standard · Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI & 20+ platform
Hooker
★ 5.1k🔥🔥 hooker is a Frida-based reverse engineering toolkit for Android. It offers a user-friendly CLI, universal scripts, auto hook generation, memory roaming to detect activities/services, one-click SOCKS5 proxy setup, Frida JustTrustMe, and BoringSSL u
Browse More Security MCP Servers
Explore all security servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Lisa Py in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Lisa Py?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.