Skip to content

MCP_Interface

MCP Interface

godon ships an MCP server (godon-mcp) as part of the Helm chart. Any MCP-compatible client — Claude Desktop, Claude Code, opencode, or any MCP SDK — can connect and manage breeders, credentials, and platform health directly.

Overview
Protocol MCP 2024-11-05 over SSE
Service godon-mcp in namespace godon
Port 3001
Endpoint http://<host>:3001/sse

The MCP server proxies to the godon API, so any action available through the REST API is also available as an MCP tool.

Available Tools

Tool Description
breeder_list List all optimization breeders with their current status
breeder_get Get detailed information about a specific breeder
breeder_create Create and start a new optimization breeder
breeder_start Resume a previously stopped breeder
breeder_stop Gracefully stop a running breeder
breeder_delete Delete a breeder and all its data
credential_list List all stored credentials
credential_create Register a new credential (SSH key, API token, etc.)
credential_get Get details of a specific credential
credential_delete Delete a stored credential
health Check platform health

Connecting

Port-forward the MCP service to your local machine:

kubectl port-forward svc/godon-mcp -n godon 3001:3001

Then point your MCP client at http://localhost:3001/sse.

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "godon": {
      "url": "http://localhost:3001/sse"
    }
  }
}

Claude Code / opencode

Add to your MCP server settings:

{
  "mcpServers": {
    "godon": {
      "url": "http://localhost:3001/sse"
    }
  }
}

Python MCP SDK

from mcp import ClientSession
from mcp.client.sse import sse_client

async with sse_client("http://localhost:3001/sse") as (read, write):
    async with ClientSession(read, write) as session:
        await session.initialize()
        tools = await session.list_tools()

TypeScript MCP SDK

import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { SSEClientTransport } from "@modelcontextprotocol/sdk/client/sse.js";

const transport = new SSEClientTransport(new URL("http://localhost:3001/sse"));
const client = new Client({ name: "my-client", version: "1.0.0" });
await client.connect(transport);
const tools = await client.listTools();

Example Usage

Once connected, an LLM can create and manage optimization runs directly. For example, to create a TCP tuning breeder:

Create a breeder named "tcp-tuning" that optimizes net.ipv4.tcp_rmem
between 4096 and 6291456 targeting host 10.0.0.5, measuring RTT as the objective.

The LLM translates this into a breeder_create tool call with the appropriate godon v0.3 configuration.