Lists the sub-task runs spawned by this execution (e.g
Connects a parent run's trace to the child runs it spawned, so the UI can navigate into each sub-task's own trace. Empty for runs that spawned no sub-tasks.
Authorization
Bearer JWT Authorization header using the Bearer scheme. Enter 'Bearer' [space] and then your token.
In: header
Path Parameters
The unique identifier of the agent.
The parent execution record identifier.
The tenant identifier
Query Parameters
The maximum number of items to return per page
int32int32Opaque cursor token from an external system (e.g. RAG API). When set, M:Fruxon.Common.Collections.PageToken.AsString returns this value directly instead of encoding Fruxon.Common.Collections.PageToken.Skip.
Response Body
curl -X GET "https://api.fruxon.com/v1/tenants/string/agents/string/executionRecords/string/children"{
"items": [
{
"id": "00000000-0000-0000-0000-000000000000",
"agentId": "string",
"agentRevision": 0,
"remoteAgentRevision": "string",
"startTime": 0,
"endTime": 0,
"status": "IN_PROGRESS",
"deliveryStatus": "NOT_APPLICABLE",
"inputTokens": 0,
"outputTokens": 0,
"cachedTokens": 0,
"cacheWriteTokens": 0,
"webSearchCalls": 0,
"hasResult": false,
"hasTrace": false,
"isSample": false,
"trigger": {
"type": "UNSPECIFIED",
"id": "string",
"provider": "string",
"triggerFireAttemptId": null,
"participantId": null,
"displayName": "string",
"participantDisplayName": "string",
"participantKind": null,
"participantAvatarFileId": "string",
"pipelineId": null
},
"destinations": [],
"environmentSlug": "string",
"networkId": "00000000-0000-0000-0000-000000000000",
"createdAt": 0,
"tenantId": "00000000-0000-0000-0000-000000000000",
"parentExecutionRecordId": "00000000-0000-0000-0000-000000000000",
"parentAgentId": "string",
"parentSubject": "string",
"parentAsyncOperationId": "00000000-0000-0000-0000-000000000000",
"rootExecutionId": "00000000-0000-0000-0000-000000000000",
"runType": "ROOT",
"subject": "string",
"inputCost": 0,
"outputCost": 0,
"cachedCost": 0,
"cacheWriteCost": 0,
"thinkingCost": 0,
"webSearchCost": 0,
"totalCost": 0,
"creditsCharged": 0,
"waitingFor": {
"kind": "string",
"participantId": null,
"displayName": "string",
"participantKind": null,
"avatarFileId": "string",
"topicId": null,
"approvalRequestId": null,
"approvers": null,
"childCount": null,
"resumesAt": null,
"attempt": null,
"maxAttempts": null,
"waitingSince": 0,
"expiresAt": null
},
"relevanceOutcome": "RESPOND",
"relevanceReason": "string",
"groupParticipantId": "00000000-0000-0000-0000-000000000000",
"wasAddressed": false,
"rejectionReason": "BUDGET_EXCEEDED",
"rejectionDetail": "string",
"rejectionAttemptCount": 0
}
],
"nextPageToken": "string",
"totalCount": 0
}{
"type": "string",
"title": "string",
"status": 0,
"detail": "string",
"instance": "string",
"property1": null,
"property2": null
}Cancels the consult a parked run is waiting on, resuming the run.
When a run is parked `WAITING_FOR_ASYNC` on a `consult_participant` to a human who hasn't answered, this lets an operator end the wait now instead of waiting out the consult's 24h TTL. The consult's proposal/topic is withdrawn and the run resumes with the same `no_answer` outcome it would get on a timeout — so cancelling introduces no new run behavior. Approvals are cancelled through their own `pendingApprovals/{id}:cancel` endpoint, not here.
Gets the summary record for a single execution.
Returns the same Fruxon.Model.ExecutionTraces.AgentExecutionRecordResponse shape every item in the list endpoint carries — start/end times, status, token counts, cost breakdown — just for one specific record id. Cheap follow-up to a streaming run: SSE emits the `executionRecordId` in the final `done` event so callers can fetch the canonical summary by id without paging through history.