What is Onos Mcp Server

The onos-mcp-server is an implementation of the Model Context Protocol (MCP) that facilitates network control and management through the ONOS SDN controller. It provides programmatic access to network management, OpenFlow device control, and analytics via ONOS’s REST API.

Use cases

Use cases for onos-mcp-server include diagnosing network connectivity issues, designing traffic flows for load balancing and security, managing network devices and links, and performing analytics on network performance metrics.

How to use

To use onos-mcp-server, developers can interact with its REST API to access various network resources, configure devices, manage flows, and perform analytics. The server supports dynamic resource access and specialized prompts for diagnosing network issues and designing network flows.

Key features

Key features of onos-mcp-server include access to over 30 dynamic resources, specialized prompts for network diagnostics and design, comprehensive analytics capabilities, and the ability to manage OpenFlow devices and network configurations.

Where to use

onos-mcp-server is suitable for educational environments, network prototyping, and SDN research. It can be utilized in scenarios where network management and control are required, particularly in software-defined networking contexts.

ONOS MCP Server

Overview

A Model Context Protocol (MCP) server implementation that provides network control and management capabilities through the ONOS SDN controller. This server enables AI assistants to access ONOS network management, OpenFlow device control, and comprehensive analytics through a structured API interface. The server is ideal for educational environments, network operations, SDN research, and AI-assisted network management.

Features

Network Resources

The server provides access to ONOS REST API endpoints, including:

• Network devices, links, and hosts
• Topology information
• Flow rules and intents
• Applications and services
• Statistics and metrics
• System and cluster health

Analytics Tools

• Network Summary: Get a comprehensive overview of devices, links, hosts, and clusters with detailed device information
• Network Analytics: Analyze performance metrics, traffic patterns, and utilization statistics
• System Health: Monitor memory usage, cluster status, and component health
• Application Management: Install, activate, deactivate, and uninstall ONOS applications
• Flow Configuration: Create and manage flow rules and intents
• Path Computation: Find optimal paths between network endpoints

Specialized Prompts

• Network Diagnostics: Troubleshoot connectivity issues and service degradation
• Intent-Based Configuration: Implement connectivity objectives and policies
• Network Health Analysis: Generate comprehensive status reports
• QoS Configuration: Set up traffic prioritization and service levels
• Performance Optimization: Improve resource allocation and scalability

Requirements

• Python 3.7+
• uv for dependency management
• Running ONOS controller
• httpx library
• mcp library

Configuration

Configure the server using environment variables:

• ONOS_API_BASE: Base URL for ONOS API (default: http://localhost:8181/onos/v1)
• ONOS_USERNAME: Username for ONOS API authentication (default: onos)
• ONOS_PASSWORD: Password for ONOS API authentication (default: rocks)

Usage with Claude Desktop

{
  "mcpServers": {
    "onos": {
      "command": "uv",
      "args": [
        "--directory",
        "parent_of_servers_repo/servers/src/onos-mcp-server",
        "run",
        "server.py"
      ],
      "env": {
        "ONOS_API_BASE": "http://localhost:8181/onos/v1",
        "ONOS_USERNAME": "onos",
        "ONOS_PASSWORD": "rocks"
      }
    }
  }
}

Contributing

Contributions are welcome! Please feel free to submit pull requests, report bugs, or suggest new features.