OPENAI-COMPATIBLE CLOUD PROXY SPECIFICATION

Cloud Proxy API Reference

Direct OpenAI-compatible endpoints with real-time token metering, sub-15ms reverse proxy routing, and zero master key leakage. Use with curl, Python, or standard OpenAI SDKs.

POST /v1/chat/completions

Streaming SSE Supported

Routes requests directly to frontier models (Claude 3.5 Sonnet, GPT-4o, DeepSeek Coder, etc.) through our sovereign reverse proxy. Master API keys are never exposed to developer machines.

# Example curl request using local client proxy token:

curl -X POST https://legecycli.onrender.com/v1/chat/completions \

-H "Authorization: Bearer cp_live_your_token_here" \

-H "Content-Type: application/json" \

-d '{

"model": "claude-3.5-sonnet",

"messages": [{"role": "user", "content": "Refactor this auth handler"}],

"stream": true

GET /api/user/balance

JSON Response

Returns current available credits, active terminal CLI sessions, and estimated token equivalents.

{
  "credits": 250,
  "approx_tokens": 1250000,
  "tier": "pro_engineer",
  "status": "active"
}

GET /v1/models

JSON Array

Lists all currently routed models available on the reverse proxy with context sizes and current token rates.