Developer Setup

Patch Real Runtime Into Existing AI Workflows

Use scoped API keys and call verification before each sensitive route. Keep your existing Gemini, GPT OS, or Claude stack as the primary runtime.

Container style

Quick Start

Run the Licensing Service

cd api_key_system
python3 -m venv .venv
source .venv/bin/activate
pip install -e .[test]
cp .env.example .env
./run.sh

Grab ready-to-run integration files and bootstrap scripts.

Download ZIP

Create a scoped test key instantly from the self-serve endpoint.

Get API Key

Patch Pattern

Call POST /v1/auth/verify with required scope and consume units.

{
  "api_key": "rk_live_...",
  "required_scope": "realflow:stream",
  "consume": true,
  "units": 1
}

If allowed, continue to your existing model endpoint with enriched context and realtime signals.

if (verify.allowed) {
  forwardToProviderRuntime(payload)
}

Endpoint Groups

Public self-serve: /v1/public/api-keys (issue test keys)

Auth: /v1/auth/verify (scope check + metering + limits)

Catalog: /v1/catalog/packages, /v1/catalog/integration-presets, /v1/catalog/addons

Usage: /v1/usage/accounts/{account_id}/summary

Demo protected: /v1/demo/realflow/ping and related routes

Recommended Scopes

realflow:stream realfuel:infer realfeel:state realcontrol:control