- Start
- Start here
Start
Start here
What Greatwait does, what a first successful result looks like, and which path to take to reach it.
Greatwait is a scheduling service. You tell it when, and at that time it sends a signed HTTPS request, a callback, to a receiver you own.
- A one-shot schedule runs once, at an instant (
at) or after a delay (delaySeconds). - A recurring schedule runs on a fixed interval (
everySeconds) or on a wall clock (cronread in atimeZone).
Each time a schedule comes due, Greatwait records an occurrence, makes one or more delivery attempts against your receiver, and keeps the outcome of every attempt so you can read it later. Each callback is signed, so your receiver can check that it came from Greatwait. Concepts explains each of these.
The production API is at https://api.greatwait.dev and the dashboard is at https://app.greatwait.dev.
What first value means
Creating a destination or a schedule only proves the API accepted your request. First value comes later, when all three of these have happened:
- A one-shot schedule you created is accepted.
- Its occurrence comes due and Greatwait makes a delivery attempt to your receiver.
- You read that attempt’s outcome, for example
state: succeededalong with the status your receiver answered.
Every quickstart below ends at step 3.
Choose your path
All five paths reach the same result. Pick the one closest to how you work.
- Website: do everything in the dashboard, with no code.
- MCP: let an AI agent create and inspect schedules through Greatwait’s MCP tools.
- curl: call the HTTP API from a shell.
- JavaScript SDK: use
@greatwait/api-clientfrom Node or any runtime withfetch. - CLI: use the
greatwaitcommand from@greatwait/cli.
Before you start, read Before you begin for what each path needs, and The invited alpha for what this release includes.
Guides
- Verifying callbacks: check signatures, handle duplicates, and choose what status to answer.
- Idempotency and conditional writes: retry safely and avoid overwriting someone else’s change.
- Recurrence and time zones: intervals, cron, and what happens when clocks change.
- Failures and retries: how delivery is retried, when it stops, and how to diagnose it.