A pipeline can refresh itself daily or weekly, so the data your models are built on stays current without anyone pressing Run. Set up the pipeline first — see Connect your warehouse.
In the pipeline builder, open Schedule in the header. Choose Daily or Weekly (and the day), then the time. Times are shown in your time zone, with the UTC hour alongside — the service schedules whole UTC hours. The popover shows when the next refresh will happen before you save.
Choose Off to pause a schedule; its settings are kept for when you turn it back on. The Schedule button always says what is set, for example “Daily · 07:00”, and the Pipelines list shows each pipeline’s schedule and how its last run went.
Needs an API key with the create:models scope.
PUT /api/v1/pipelines/{pipeline_ref}/schedule
{"cadence": "weekly", "hour_utc": 6, "weekday": 0, "enabled": true}
GET /api/v1/pipelines/{pipeline_ref}/schedule
→ {"cadence": "weekly", "hour_utc": 6, "weekday": 0, "enabled": true, "next_run_at": "2026-10-05T06:00:00"}
Over MCP:
set_pipeline_schedule(pipeline_ref="ab12cd34ef", cadence="weekly", hour_utc=6, weekday=0, enabled=True)
| Field | Values |
|---|---|
cadence |
daily or weekly |
hour_utc |
A whole hour, 0–23, in UTC |
weekday |
0 = Monday … 6 = Sunday; weekly only (omit it for daily) |
enabled |
false pauses the schedule and keeps its settings |
next_run_at is in UTC. A pipeline that has never been scheduled returns enabled: false and nulls.
get_pipeline_run). It runs as the pipeline’s owner, with the pipeline’s saved connections.