mcpField
O
omni-providers

omni-providers

github.com/cheaper-inferenceCloud & DevOps82.3k1.2k installsUpdated today
Repository

About omni-providers

omni-providers is a skill for managing AI provider connections through the OmniRoute REST API. It supports connection lifecycle operations, authentication flows, testing, model management, and health and usage inspection.

Key features

  1. 01List, add, update, and remove connections
  2. 02Manage API keys and OAuth flows
  3. 03Test connections and batches
  4. 04Import and refresh credentials
  5. 05Inspect models, metrics, and health

Use cases

  1. 01Add or remove a provider connection
  2. 02Renew credentials after token expiry
  3. 03Test providers before routing
  4. 04Review provider health and usage

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/cheaper-inference/omniroute --skill omni-providers

After installing, type/omni-providersin Claude Code to confirm it loaded.

Bundled filesreferences/endpoints.md

Details

From the project README

Overview

Manage provider connections, API keys, OAuth flows, and connection tests via the REST API. List, add, update, remove, and test AI provider integrations across OmniRoute's 375-provider catalog.

Authentication

All requests require a valid Bearer token or session cookie. Obtain a token via POST /api/auth/login or configure REQUIRE_API_KEY=false for local development.

Endpoints

Payloads

See the full OpenAPI specification at GET /api/openapi/spec or docs/openapi.yaml for detailed request/response schemas.

Additional endpoints

POST /api/providers/{id}/refresh

Rotate / refresh OAuth tokens or re-validate credentials for a provider connection. Useful after token expiration or when credentials need to be renewed without deleting the connection.

curl -X POST https://localhost:20128/api/providers/{id}/refresh \
  -H "Authorization: Bearer $OMNIROUTE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{}'
GET /api/provider-metrics

Retrieve aggregated usage and performance metrics for all provider connections (request counts, error rates, latency percentiles, token usage). Used by dashboards and Auto-Combo scoring.

curl https://localhost:20128/api/provider-metrics \
  -H "Authorization: Bearer $OMNIROUTE_TOKEN"

Response example:

{
  "providers": [
    {
      "id": "my-openai",
      "requestCount": 1240,
      "errorRate": 0.02,
      "p50LatencyMs": 340,
      "p99LatencyMs": 1200,
      "tokensIn": 500000,
      "tokensOut": 250000
    }
  ]
}
GET /api/providers/health-matrix

Returns a health-matrix view of all providers with per-connection status, circuit-breaker state, and cooldown expiry. Useful for debugging routing decisions.

curl https://localhost:20128/api/providers/health-matrix \
  -H "Authorization: Bearer $OMNIROUTE_TOKEN"

Comments

Markdown supported