MCPSERV.CLUB
MCP-Mirror

ONOS MCP Server

MCP Server

Programmatic ONOS control and analytics via Model Context Protocol

Stale(50)
0stars
2views
Updated Apr 3, 2025

About

The ONOS MCP Server exposes a rich set of resources and prompts that enable developers, researchers, and educators to programmatically manage OpenFlow devices, design flows, monitor system health, and analyze metrics through the ONOS SDN controller. It supports network prototyping, migration planning, and performance tuning.

Capabilities

Resources
Access data sources
Tools
Execute functions
Prompts
Pre-built templates
Sampling
AI model interactions

ONOS MCP Server Overview

The ONOS MCP Server is a purpose‑built bridge between AI assistants and the ONOS SDN controller, turning complex network management tasks into simple, declarative queries. By exposing a rich set of dynamic resources—devices, links, hosts, topology, flows, intents, and system metrics—the server gives an AI the same programmatic view that a network engineer would normally access through ONOS’s REST API. This removes the need for custom integrations or manual parsing of raw JSON, allowing conversational agents to retrieve real‑time network state with a single request.

Developers benefit from the server’s nine specialized prompts, each tailored to a common SDN workflow. Whether diagnosing connectivity problems, designing traffic isolation policies, or planning a migration from legacy switches to an intent‑driven architecture, the prompts translate natural language into precise ONOS queries and actionable recommendations. For example, a “network‑health‑report” prompt aggregates device availability, flow statistics, and bottleneck alerts into a concise status summary that can be fed directly into an AI‑powered dashboard or chatbot.

Beyond passive data retrieval, the server equips AI assistants with over twenty powerful tools that perform stateful operations. Tools such as or encapsulate complex sequences of REST calls into single, well‑defined actions. This abstraction enables developers to script routine tasks—like installing a new application or retrieving system health—in an intuitive, intent‑driven manner. The result is tighter integration between AI workflows and the ONOS control plane, reducing boilerplate code and accelerating prototype development.

In real‑world scenarios, the ONOS MCP Server shines in educational labs, research projects, and production environments where rapid experimentation is essential. Students can interact with a live SDN topology through natural language queries, while researchers can automate performance tuning and anomaly detection. Network operators can embed the server into chatops platforms, allowing instant troubleshooting or policy deployment without leaving their conversational interface. By unifying network visibility and control under the MCP umbrella, this server provides a scalable, developer‑friendly gateway to ONOS’s full capabilities.