Installation

mcpcap requires Python 3.10 or greater.

Using uv

If you’re using uv:

uv add mcpcap

Using uvx (One-time usage)

To run mcpcap without installing it permanently:

uvx mcpcap

Then connect with an MCP client and pass a PCAP path or URL to the analysis tool you want to call.

For HTTP-based MCP clients, start mcpcap with:

mcpcap --transport http --host 127.0.0.1 --port 8080

Development Installation

If you want to contribute to mcpcap or modify it:

# Clone the repository
git clone https://github.com/mcpcap/mcpcap.git
cd mcpcap

# Install in development mode with all dependencies
pip install -e ".[dev,docs,test]"

# Install the repository's Git hooks
pre-commit install

If you prefer uv for development:

uv sync --extra dev --extra docs --extra test
pre-commit install

Verify Installation

Verify that mcpcap is installed correctly:

mcpcap --help

You should see the help message showing available command-line options.

Dependencies

mcpcap depends on:

  • fastmcp: MCP server framework

  • scapy: Packet parsing and analysis

  • requests: HTTP client for remote PCAP access

  • Python 3.10+: Modern Python features and type hints

All dependencies are automatically installed when you install mcpcap.

Troubleshooting

Permission Issues

mcpcap analyzes existing capture files. It does not capture live traffic itself, so elevated packet-capture privileges are not normally required.

Import Errors

If you encounter import errors, make sure you’re using the correct Python version:

python --version  # Should be 3.10 or higher

Virtual Environment

It’s recommended to use a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install mcpcap