Status

Uteke Cloud does not have a separate status page yet. You can check the service yourself in a few seconds, and this page says where we report problems.

Is it up?

Ask the public health endpoint. It needs no account:

curl https://uteke.app/health

A running service answers uteke-cloud <version> ok. No answer, or any other answer, means the service is down or restarting.

Is my workspace healthy?

With an API key from your portal, ask for the health of each part:

curl https://uteke.app/api/v2/health \
  -H "Authorization: Bearer utk_live_YOUR_KEY"

The answer reports the database, the vector store, the embedding service and the search index. The status is degraded when one of them is not working, for example when new memories wait too long to be indexed for search. Those memories are still stored, and keyword search finds them in the meantime.

Where incidents are reported

• Fixes for incidents are listed in the changelog with the release that contains them.
• If a problem affects your personal data, we notify you by email as the privacy policy describes.
• Something looks wrong and the health endpoint says ok? Email [email protected] with what you saw and when.

If the service is down

Your memories are stored in the database, and the search index can be rebuilt from it in full, so an outage does not lose memories. Once the service is back, your agents recall as before. You can keep a copy of everything at any time with an export.