sec

sec-mcp is a Python-based MCP server that exposes the sec-mcp threat intelligence checks to an LLM environment. It loads the same in-memory, high-performance threat feeds used by the CLI and Python API, allowing your language model workflows to enrich responses with real-time domain, URL, and IP reputation data. The server supports a variety of tools for clients to query, update, and diagnose the threat data store, including batch checks, status reporting, blacklist updates, and diagnostics. You can enable fast, v2-storage mode via the MCP_USE_V2_STORAGE flag to optimize performance in production deployments.

To use the MCP server, configure an MCP client (for example a Claude, Windsurf, or Cursor integration) to point at the server entry point and pass the appropriate Python environment path. The available tools include batch checks, status queries, updates, diagnostics, and manual entry management. The server is designed to minimize token usage in communications while preserving full functionality, and it leverages an in-memory storage backend with optional v2 features for speed and scalability.

Prerequisites:

• Python 3.8+ installed on the host
• A working virtual environment (recommended)
• Network access to install packages from PyPI

Step-by-step installation:

1. Create and activate a virtual environment (optional but recommended):

python3 -m venv .venv
source .venv/bin/activate  # macOS/Linux
.\.venv\Scripts\activate.bat  # Windows

2. Install the sec-mcp package from PyPI:

pip install sec-mcp

3. (Optional) Initialize or update blacklists if your workflow requires a local cache:

sec-mcp update

4. Run the MCP server in a mode suitable for your environment (example shown below):

python -m sec_mcp.start_server

5. In your MCP client configuration, point to the server entry, for example using an absolute Python path in a virtual environment:

{
  "mcpServers": {
    "sec-mcp": {
      "command": "/absolute/path/to/.venv/bin/python",
      "args": ["-m", "sec_mcp.start_server"],
      "env": {
        "MCP_USE_V2_STORAGE": "true"
      }
    }
  }
}

Environment variables and configuration:

• MCP_USE_V2_STORAGE: Enable v2 in-memory storage for higher performance. Set to true to activate (requires compatible Python and storage backend).
• Ensure the absolute path to your Python executable is provided in MCP client configurations when launching the MCP server from a virtual environment.
• For production deployments, consider running the server behind a secure gateway or load balancer and restricting access to trusted clients.
• If you encounter missing dependencies, ensure pip is up-to-date and that the environment has network access to PyPI during installation.