Pyats MCP

🚀 pyATS MCP Server

This project implements a server based on the Model Context Protocol (MCP), which encapsulates the capabilities of Cisco pyATS and Genie. It enables structured, model-based interactions with network devices using the JSON-RPC 2.0 protocol over STDIO.

🚨 This server does not use HTTP or SSE. All communications are done via standard input/output (STDIN/STDOUT), making it an ideal choice for secure embedded, containerized, or LangGraph tool integrations.

✨ Features

• Connect to Cisco IOS/NX-OS devices with device information from the pyATS testbed.
• Support the secure execution of verified CLI commands (show, ping).
• Allow controlled configuration changes.
• Return structured (parsed) or raw output.
• Expose a well-defined set of tools through tools/discover and tools/call.
• Run entirely over STDIO for a minimal attack surface and maximum portability.

🚀 Quick Start

Set up your testbed path

export PYATS_TESTBED_PATH=/absolute/path/to/testbed.yaml

Start the server

Continuous STDIO mode (default)

python3 pyats_mcp_server.py

Starts a long-running process that reads JSON-RPC requests from standard input and writes to standard output.

One-shot mode

echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/discover"}' | python3 pyats_mcp_server.py --oneshot

Processes a single JSON-RPC request and then exits.

📦 Docker Support

Build the container

docker build -t pyats-mcp-server .

Run the container (STDIO mode)

docker run -i --rm \
  -e PYATS_TESTBED_PATH=/app/testbed.yaml \
  -v /your/testbed/folder:/app \
  pyats-mcp-server

💻 Usage Examples

Available MCP Tools

All inputs are validated through Pydantic models to ensure security and consistency.

LangGraph Integration

Add the MCP server as a tool node to your LangGraph pipeline as follows:

("pyats-mcp", ["python3", "pyats_mcp_server.py", "--oneshot"],)

📄 Author

John Capobianco Product Marketing Evangelist at Selector AI Author of Automate Your Network

If you have any questions, feel free to let me know. I'd be happy to add:

• A sample LangGraph graph configuration.
• A companion client script.
• CI/CD integration (e.g., GitHub Actions).

I'm glad to help!

⚠️ Important Note

The testbed.yaml file is compatible with Cisco Modeling Labs (CML) sandboxes on Cisco DevNet!