The Get started checklist
New projects show a Get started checklist on their overview. It exists to make the model concrete: an engine you wake, a key you issue once, a data plane you probe, and services you exercise.
| # | Step | Detected by |
|---|---|---|
| 1 | Wake up the engine | project.state === "running" |
| 2 | Create a project key and copy it once | at least one non-revoked key in GET …/keys |
| 3 | Probe /v2/capabilities with the key |
a local flag set by the Engine page after a 200 |
| 4 | Create a bucket and upload an object | at least one bucket in GET …/storage/buckets |
| 5 | Publish a realtime event | a local flag set by the Realtime page after a successful publish |
| 6 | Deploy and invoke a function | at least one function in GET …/functions |
| 7 | Sleep, then create a snapshot | at least one snapshot in GET …/snapshots |
Steps 3 and 5 cannot be read from any route (a probe and a publish leave no trace by design), so they are remembered in this browser only. Everything else is derived from the same responses the pages already fetch. Dismissing the checklist hides it for this project in this browser.