OngoingAI Gateway

Use this documentation set to deploy and operate OngoingAI Gateway. Use it as a progressive path from first request routing to production policy and operational troubleshooting.

Use this documentation when

You are setting up OngoingAI Gateway for the first time.
You need architecture and policy details for production rollout.
You need exact reference material for config, CLI, and API integration.

How to read these docs

Get started to install the gateway, route first traffic, and validate traces.
Core concepts to understand request flow, trace behavior, privacy model, and tenant scope.
Capabilities to configure auth, routing, tracing, analytics, limits, and storage.
Reference for exact config keys, CLI commands, env overrides, and API contracts.
Help for troubleshooting, security guidance, FAQ, and contribution workflows.

If you already run the gateway in production, start at Reference and Help.

Audience

Engineers integrating SDK and CLI traffic through the gateway.
Platform and security teams operating multi-tenant deployments.
Contributors extending provider, storage, and policy behavior.

Product scope

OngoingAI Gateway is a headless AI gateway. Web UI and end-user session flows are out of scope for this repository.

Releases