Status and health checks

Right now

The check above runs in your browser with a read-only GET https://api.dualmindlab.tech/health. If it fails here but works with curl, suspect your network or an extension first.

Health endpoints

Endpoint	Auth	Returns	Use it for
`GET /health`	None	`status`, `message`, `version`, `timestamp`, `instanceId`	Liveness, the cron keep-alive and the deploy check
`GET /api/health`	None	Same as `/health`	Clients that only proxy `/api/*`
`GET /api/arena/ping`	None	`success`, `message`, `version`	A route through the arena controller
`GET /api/admin/dashboard/health`	AdminOnly	`status`, `supabase_connected`	Confirms the API can reach the database (and that you’re an admin)
`GET /api/arena/model-stats`	None	Leaderboard rows	A read through Supabase and the `v_leaderboard` view

/health answers from the container without touching Supabase or any provider. A healthy /health with failing battles points at configuration (Supabase keys, AI Gateway) rather than hosting. See Troubleshooting.

curl -s https://api.dualmindlab.tech/health

const res = await fetch("https://api.dualmindlab.tech/health", { cache: "no-store" });
const { status, version, instanceId } = await res.json();
console.log(status, version, instanceId);

import requests

health = requests.get("https://api.dualmindlab.tech/health", timeout=10).json()
print(health["status"], health["version"], health["instanceId"])

Response

{
  "status": "healthy",
  "message": "DualMind API is running",
  "timestamp": "2026-10-17T19:54:46.7135698Z",
  "version": "1.0.0",
  "instanceId": "67c5fc5b5352c545…"
}

instanceId is the Durable Object ID of the container that answered (local when you run the API yourself). With one container it stays the same until the next deploy or restart.

A live read through the database

The leaderboard below is fetched from your browser with GET https://api.dualmindlab.tech/api/arena/model-stats. If it loads, the API, Supabase and the v_leaderboard view are all working.

Watching the service

Keep-alive

The dualmind-api Worker’s cron pings /health every five minutes; its log line is [keep-alive] Container health: 200.

Logs

npx wrangler tail dualmind-api streams Worker and container logs. Every request logs a start and end line with its correlation ID, status and duration.

After a deploy

CI polls /health up to five times. Then run the release checklist.

There is no public status page or external uptime monitor wired up yet. Adding one (and alerts on error rates) is on the roadmap.

Get started

Concepts

Architecture

Data

Develop

Deploy and operate

Right now

Health endpoints

A live read through the database

Watching the service

​Right now

​Health endpoints

​A live read through the database

​Watching the service

Right now

Health endpoints

A live read through the database

Watching the service