pyATS_MCP

pyATS_MCP

3.5

If you are the rightful owner of pyATS_MCP and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The pyATS MCP Server is a Model Context Protocol Server that integrates with Cisco pyATS and Genie, enabling structured interaction with network devices using JSON-RPC 2.0 over STDIO.

The pyATS MCP Server is designed to facilitate secure and structured communication with Cisco network devices by leveraging the pyATS and Genie frameworks. It operates using the JSON-RPC 2.0 protocol over standard input/output (STDIO), making it suitable for secure, embedded, and containerized environments. The server connects to devices defined in a pyATS testbed, allowing for the execution of validated CLI commands, controlled configuration changes, and retrieval of structured or raw outputs. It provides a set of tools for network operations, all while ensuring minimal surface area and maximum portability. The server's STDIO-based communication model is particularly advantageous for integration with LangGraph pipelines, as it avoids the need for HTTP ports or REST endpoints. Security is a key focus, with input validation and command restrictions in place to prevent unsafe operations.

Features

  • Secure STDIO Communication: Operates entirely over STDIN/STDOUT, enhancing security and portability.
  • Validated Command Execution: Supports safe execution of CLI commands with input validation using Pydantic.
  • Structured Output: Returns both parsed and raw outputs for network operations.
  • Docker Support: Easily containerized for deployment in various environments.
  • LangGraph Integration: Seamlessly integrates with LangGraph pipelines for tool invocation.

Tools

  1. run_show_command

    Safely execute show commands, select to parse output

  2. run_ping_command

    Execute a ping test and return parsed or original results

  3. apply_configuration

    Application-safe configuration commands (support multiple lines)

  4. learn_config

    Get the run brief

  5. learn_logging

    Get the system log (show logging last 250)