Goose

Goose is an open-source, on-machine AI agent — desktop app, CLI, and API — governed by the Agentic AI Foundation (AAIF) at the Linux Foundation. It speaks to any OpenAI-compatible endpoint, so you can add Router9 as a custom provider without any code changes.

Setup

Goose reads custom providers from JSON files under ~/.config/goose/custom_providers/ (on Windows, %APPDATA%\Block\goose\config\custom_providers\). Create router9.json there:

{
  "name": "router9",
  "engine": "openai",
  "display_name": "Router9",
  "description": "Router9 flat-rate LLM gateway",
  "api_key_env": "ROUTER9_API_KEY",
  "base_url": "https://api.router9.com/v1/chat/completions",
  "models": [
    { "name": "auto", "context_limit": 200000 },
    { "name": "claude-sonnet-4-20250514", "context_limit": 200000 }
  ],
  "supports_streaming": true,
  "requires_auth": true
}

You can also run goose configure, choose Custom Providers → Add A Custom Provider, pick API Type OpenAI Compatible, and enter https://api.router9.com/v1/chat/completions as the API URL.

API Key

Provide your Router9 key through the environment variable named in api_key_env:

export ROUTER9_API_KEY=sk-r9k-your-key-here

Then start a session on the provider:

goose session start --provider router9

Choosing Models

The models array lists the ids Router9 routes on — select one at session start or via goose configure. Use auto to let Router9 pick the best model for your plan, or pin any model Router9 supports:

{ "name": "gpt-4o", "context_limit": 128000 }

Add Skills via MCP

Goose is built around extensions (MCP servers). Add Router9's hosted MCP server as a remote extension to expose Router9 Skills as tools — see Install MCP Server.

Tips

• Router9's flat monthly pricing suits Goose's long, autonomous sessions — no per-token surprises.
• Keep ROUTER9_API_KEY in your shell environment rather than committing it to the provider JSON.