Settles a follow-up step whose outcome is UNKNOWN, once you have checked whether it landed.
A hook or a writing agent step whose runner stopped mid-step is never re-run blindly. Say `landed: true` if it happened; `false` and it is tried again while it has attempts left (`nextAttemptAt` is set). One that has had them all stays FAILED, and `:rerunFollowUp` tries it again. Either way, once every step has come to rest the cohort completes at once.
Authorization
Bearer JWT Authorization header using the Bearer scheme. Enter 'Bearer' [space] and then your token.
In: header
Path Parameters
The trigger identifier.
uuidThe cohort identifier.
uuidThe follow-up step identifier.
uuidThe tenant identifier
Whether it landed.
True: it happened. False: it did not, and may be tried again.
Response Body
curl -X POST "https://api.fruxon.com/v1/tenants/string/triggers/497f6eca-6276-4993-bfeb-53cbbbba6f08/ledger/cohorts/497f6eca-6276-4993-bfeb-53cbbbba6f08/followUps/497f6eca-6276-4993-bfeb-53cbbbba6f08:settle" \ -H "Content-Type: application/json" \ -d '{}'{
"id": "00000000-0000-0000-0000-000000000000",
"generation": 0,
"stepIndex": 0,
"kind": "NOTIFY",
"step": {
"kind": "NOTIFY",
"audience": "OPENER",
"hook": "string",
"agentId": "string",
"instruction": "string",
"actionCeiling": "READ_ONLY",
"after": 0,
"effectiveActionCeiling": "READ_ONLY",
"isSafeToRepeat": true
},
"status": "PENDING",
"attempts": 0,
"nextAttemptAt": 0,
"detail": "string",
"externalRef": "string",
"executionRecordId": "00000000-0000-0000-0000-000000000000",
"resolvedAt": 0
}{
"type": "string",
"title": "string",
"status": 0,
"detail": "string",
"instance": "string",
"property1": null,
"property2": null
}{
"type": "string",
"title": "string",
"status": 0,
"detail": "string",
"instance": "string",
"property1": null,
"property2": null
}Confirms a sampled cohort's estimate, and lets the rest of it run.
A sampled cohort ran its first items for real and admits nothing more until someone confirms what the rest is projected to cost. Confirming resumes admission, raises the platform-managed limits its runs stop at to cover the projection (put back when the cohort settles), lets a paused automation work through it, and starts a pass now. Those limits are the automation's own when it has an enforced one (`estimate.automationBudgetImpact`), and otherwise those of the agents it runs on (`estimate.budgetImpact`); limits a person owns are never raised. Pass the `estimate.hash` you were shown; 409 if the estimate has changed since, or the cohort is not awaiting confirmation.
Runs the cohort's follow-up again, over its counts as they are now.
After a redrive fixed what was stuck, or a step failed for good. A new generation of steps; the old ones stay as the record of what ran. 409 while a follow-up is still running.