generated: '2026-07-20'
method: searched
source: https://docs.openmind.com/api-reference + https://docs.openmind.com/full-autonomy-guidelines/api_endpoints
docs: https://docs.openmind.com/api-reference/introduction
conventions:
  authentication:
    style: api-key (header x-api-key or Authorization Bearer) for LLM/speech/vision; Clerk JWT bearer for Core account API
    ref: authentication/openmind-authentication.yml
  base_urls:
    cloud: https://api.openmind.com
    core: https://api.openmind.com/api/core
    websocket: wss://api.openmind.com
    premium_orchestrator: http://<robot-host>:5000  # OM1 ROS2 SDK, runs on-robot
    premium_nav2: http://<robot-host>:5001          # OM1 ROS2 SDK, runs on-robot
  llm_interface:
    compatibility: OpenAI Chat Completions API format
    path: /api/core/{provider}/chat/completions
    providers: [openai, deepseek, gemini, xai, nearai, openrouter]
  metering:
    unit: OMCU (OpenMind Computational Unit)
    balance_endpoint: GET /api/core/account/balance
    notes: monthly credits reset each billing period; unused prepaid credits carry over
  idempotency:
    supported: false
    notes: no idempotency key/header documented in the OpenMind API reference
  pagination:
    supported: false
    notes: list endpoints (e.g. GET /api_keys) return full arrays; no pagination documented
  error_envelope:
    shape: '{"error": "<message>"}'
    format: custom-json (not RFC 9457 problem+json)
    ref: errors/openmind-problem-types.yml
  rate_limiting:
    style: per-plan requests/second ceiling; 429 Too Many Requests on exceed
    ref: rate-limits/openmind-rate-limits.yml
  content_type: application/json
  versioning:
    scheme: path-prefixed (/api/core); no dated/header API version documented
    ref: lifecycle/openmind-lifecycle.yml