Skip to main content
Coming in the first release. This page describes behaviour being built for the first Gliss release. Details may change before it ships.
After each push, your AI waits for the release and receives a diagnosis:
  • Status: healthy, or failing.
  • Whose problem it is: the app’s, or the platform’s.
  • A one-line summary, for example: the migration failed with relation "users" already exists.
  • The next step, and who should take it: the app owner, a maintainer, or IT.
When it is the app’s problem, your AI fixes the code and pushes again. When it is the platform’s, the diagnosis says so, and your AI should stop trying: see Troubleshooting. Your AI can also read logs, events and status for each stage. Logs are kept in your company’s account, so they survive restarts, and they are redacted before they leave it. Your IT team decides whether diagnostics may leave the account at all.