self-setup

Self-Setup — Provider & Model Configuration

This skill lets you configure LLM providers and models through conversation. You have access to your own management API at http://localhost:8088/api/.

---

When to Use

• User asks to add, change, or configure an LLM provider
• User asks to switch models
• User provides an API key for a provider
• User asks "what models are available?"
• User says something like "use GPT-5" or "switch to Claude"

---

API Reference

All commands use curl via your shell tool. Base URL: http://localhost:8088/api

List All Providers

curl -s http://localhost:8088/api/models | python3 -m json.tool

Get Active Model

curl -s http://localhost:8088/api/models/active | python3 -m json.tool

Set Active Model

curl -s -X PUT http://localhost:8088/api/models/active \
  -H "Content-Type: application/json" \
  -d '{"provider_id": "PROVIDER_ID", "model": "MODEL_ID"}'

Configure Provider API Key

curl -s -X PUT http://localhost:8088/api/models/PROVIDER_ID/config \
  -H "Content-Type: application/json" \
  -d '{"api_key": "USER_API_KEY"}'

Add Custom Provider

For providers not in the built-in list (Together AI, Groq, Fireworks, DeepInfra, etc.):

curl -s -X POST http://localhost:8088/api/models/custom-providers \
  -H "Content-Type: application/json" \
  -d '{
    "id": "provider-id",
    "name": "Provider Name",
    "default_base_url": "https://api.example.com/v1",
    "api_key_prefix": "sk-",
    "models": [
      {"id": "model-id", "name": "Model Display Name"}
    ]
  }'

Test Provider Connection

curl -s -X POST http://localhost:8088/api/models/PROVIDER_ID/test \
  -H "Content-Type: application/json" \
  -d '{"api_key": "OPTIONAL_KEY", "base_url": "OPTIONAL_URL"}'

Delete Custom Provider

curl -s -X DELETE http://localhost:8088/api/models/PROVIDER_ID

Add Model to Existing Provider

curl -s -X POST http://localhost:8088/api/models/PROVIDER_ID/models \
  -H "Content-Type: application/json" \
  -d '{"id": "model-id", "name": "Model Name"}'