Delete Pipeline

Delete a pipeline run and its owned data: value cells, record sets, and cascaded documents and field states. The seed record set is untouched.

Delete a pipeline run and the data it owns. Deletion removes the run's value cells, its owned record sets, and cascades to its pipeline documents, field states, validation results, and review decisions. The Spec's seed record set (the shared registry the run read from) is not touched, so deleting a run never damages the source data other runs depend on.

Deletion runs inside a transaction so the run and its owned data are removed atomically. This is a destructive, irreversible operation: the extracted cells, the compiled phase config, and the per-document state for this run are gone. If you produced a data product from the run, manage that separately through the data-products endpoints.

The run is tenant-scoped: a run that belongs to another organization returns 404 not_found. The master view (all) can read runs but cannot delete them, since deletion is a tenant-scoped mutation. The response confirms the deletion with the deleted run's id. This endpoint requires an API key with the write scope.

Response

{
  "deleted": true,
  "id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890"
}

Errors