guides/ai-assistant.md
AI Assistant
Mino runs the assistant server-side and exposes it via POST /api/v1/agent/chat.
Configure Agent
Set env vars (or config) on the server:
MINO_AGENT_ENABLED=true
MINO_AGENT_PROVIDER=openai
MINO_AGENT_API_KEY=sk-...
MINO_AGENT_MODEL=gpt-4o-mini
Check Agent Status
curl -H "X-Mino-Key: <API_KEY>" \
http://localhost:3000/api/v1/agent/status
Chat Endpoint
curl -X POST \
-H "X-Mino-Key: <API_KEY>" \
-H "Content-Type: application/json" \
-d '{"message":"Summarize notes about launch prep"}' \
http://localhost:3000/api/v1/agent/chat
Notes
- Agent requests are authenticated with
X-Mino-Key - Tool execution depends on current plugin/runtime state
- Keep prompts explicit and include note path/folder context when possible