AI Gateway
Transparency

A lightweight OpenAI-compatible gateway surface for routing, runtime evidence, and token truth. This console shows the current mock-only deployment without exposing prompts, headers, or provider secrets.

Checking runtime
Gateway mode
-
Readiness is checked without paid provider calls.
Provider call
-
Must remain false for this mock-only baseline.
Token truth
-
Mock mode uses runtime estimates, not billing truth.
Codex routed
false
Codex remains the controller, not a routed client.

Quick Start

OpenAI-compatible 
from openai import OpenAI

client = OpenAI(
    base_url="https://api.omni.doloclaw.com/v1",
    api_key="YOUR_OMNI_KEY"
)

response = client.chat.completions.create(
    model="omni-runtime-mock",
    messages=[{"role": "user", "content": "Hello Omni"}]
)

Runtime Path

no trace yet
Clientwaiting
Gatewaymock-only
Providernot called
Reportmetadata-only