Queue an operator response to the latest Agent conversation action

POST

api

agents

runs

{agentRunId}

actions

{actionId}

resolve

Queue an operator response to the latest Agent conversation action

curl --request POST \
  --url https://api.revoengine.com/api/v1/agents/runs/{agentRunId}/actions/{actionId}/resolve \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{}'

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({})
};

fetch('https://api.revoengine.com/api/v1/agents/runs/{agentRunId}/actions/{actionId}/resolve', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

import requests

url = "https://api.revoengine.com/api/v1/agents/runs/{agentRunId}/actions/{actionId}/resolve"

payload = {}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

{
  "agentRunId": "aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa",
  "agentId": "99999999-9999-4999-8999-999999999999",
  "agentSessionId": "abababab-abab-4bab-8bab-abababababab",
  "status": "PROCESSING",
  "serviceAccountUserId": "11111111-1111-4111-8111-111111111111",
  "tickSeq": 3,
  "tickCount": 2,
  "childCount": 1,
  "state": {
    "instruction": "Review deployment"
  },
  "createdAt": "2026-06-13T10:00:00.000Z",
  "lastUpdateAt": "2026-06-13T10:15:00.000Z",
  "version": 1,
  "previousRunId": "aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa",
  "rootRunId": "aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa",
  "parentRunId": "bbbbbbbb-bbbb-4bbb-8bbb-bbbbbbbbbbbb",
  "waitReason": "CHILDREN",
  "triggerType": "REQUEST",
  "triggerRefType": "scheduleId",
  "triggerRefId": "schedule-123",
  "ownerId": "22222222-2222-4222-8222-222222222222",
  "ownerType": "USER",
  "rootAssistantThreadId": "cccccccc-cccc-4ccc-8ccc-cccccccccccc",
  "currentAssistantThreadId": "dddddddd-dddd-4ddd-8ddd-dddddddddddd",
  "lastAssistantMessageId": "eeeeeeee-eeee-4eee-8eee-eeeeeeeeeeee",
  "nextWakeAt": "2026-06-13T10:30:00.000Z",
  "scheduledFor": "2026-06-13T10:30:00.000Z",
  "lastHeartbeatAt": "2026-06-13T10:20:00.000Z",
  "checkpointSummary": "Run completed and wrote the release summary.",
  "result": {
    "summary": "Deployment looks healthy."
  },
  "error": {
    "message": "Run failed."
  }
}

Authorizations

Authorization

string

header

required

Bearer JWT token or API Key (sk-*).

Path Parameters

agentRunId

string

required

actionId

string

required

Body

application/json

decision

enum<string>

required

Resolution decision for the pending action.

Available options:

approve,

approve_all_for_turn,

reject,

submit

storageEntryIds

string<uuid>[]

Finalized Storage files linked to this action response.

Maximum array length: 8

actionVersion

number

Optimistic action version used as the idempotency key together with the action id.

Required range: x >= 0

actionItemId

string

Optional approval item id when resolving one item from a multi-approval action.

Maximum string length: 128

responseText

string

Optional free-text response for clarification or correction context.

Maximum string length: 4000

responsePayload

object

Optional structured payload for clarification answers or action metadata.

backgroundProcessing

boolean

Allow the Assistant turn to continue after the client disconnects. Defaults to false unless the operation states otherwise.

model

string

Optional execution model identifier enabled for this instance.

fastMode

boolean

Optional Fast-mode override for the resumed turn. Available only on GPT-6 Luna.

reasoningEffort

enum<string>

Optional assistant reasoning effort override for the resumed turn: none, low, medium, high, or xhigh.

Available options:

none,

low,

medium,

high,

xhigh

reasoning

enum<string>

Alias for reasoningEffort used by assistant preferences. Prefer reasoningEffort for new clients.

Available options:

none,

low,

medium,

high,

xhigh

executionMode

enum<string>

Execution mode override for the resumed turn after this action is resolved.

Available options:

interactive,

execute,

background,

automation

planningPolicy

enum<string>

Planning policy for the turn. standard can create and show progress; review_required creates a user-reviewed plan before execution.

Available options:

standard,

review_required

toolNames

string[]

Optional operational tool names admitted for the resumed turn. Omit to keep the current thread admission; pass an empty array to admit no tools for direct-answer continuation.

pluginIds

string<uuid>[]

Optional additional registry plugin ids attached to the resumed turn.

Response

202 - application/json

agentRunId

string

required

Stable run id.

Example:

"aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa"

agentId

string

required

Owning agent id.

Example:

"99999999-9999-4999-8999-999999999999"

agentSessionId

string

required

Durable session shared by successor runs.

Example:

"abababab-abab-4bab-8bab-abababababab"

status

enum<string>

required

Current run status.

Available options:

CREATED,

PENDING,

PROCESSING,

ACTION_REQUIRED,

FINISHED,

ERROR,

CANCELLED

Example:

"PROCESSING"

serviceAccountUserId

string

required

Service account execution principal.

Example:

"11111111-1111-4111-8111-111111111111"

tickSeq

number

required

Current tick sequence.

Example:

3

tickCount

number

required

Number of ticks executed.

Example:

2

childCount

number

required

Number of child runs created.

Example:

1

state

object

required

Run-scoped state for operator diagnostics.

Example:

{ "instruction": "Review deployment" }

createdAt

string<date-time>

required

Creation timestamp.

Example:

"2026-06-13T10:00:00.000Z"

lastUpdateAt

string<date-time>

required

Last update timestamp.

Example:

"2026-06-13T10:15:00.000Z"

version

number

required

Optimistic locking version.

Example:

1

previousRunId

string

Immediately preceding run in this session.

Example:

"aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa"

rootRunId

string

Root run id for lineage.

Example:

"aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa"

parentRunId

string

Parent run id for delegated child runs.

Example:

"bbbbbbbb-bbbb-4bbb-8bbb-bbbbbbbbbbbb"

waitReason

enum<string>

Reason a run is waiting.

Available options:

NONE,

CHILDREN,

CAPACITY,

SCHEDULE,

RECONCILE

Example:

"CHILDREN"

triggerType

string

Trigger type such as REQUEST, INBOX, SCHEDULE, EVENT, or RETRY.

Example:

"REQUEST"

triggerRefType

string

External trigger reference type.

Example:

"scheduleId"

triggerRefId

string

External trigger reference id.

Example:

"schedule-123"

ownerId

string

Owning user or group id.

Example:

"22222222-2222-4222-8222-222222222222"

ownerType

string

Owner type, typically USER or GROUP.

Example:

"USER"

rootAssistantThreadId

string

Root assistant thread for the run.

Example:

"cccccccc-cccc-4ccc-8ccc-cccccccccccc"

currentAssistantThreadId

string

Current assistant thread for the run.

Example:

"dddddddd-dddd-4ddd-8ddd-dddddddddddd"

lastAssistantMessageId

string

Last assistant message id.

Example:

"eeeeeeee-eeee-4eee-8eee-eeeeeeeeeeee"

nextWakeAt

string<date-time>

Next scheduled wake timestamp.

Example:

"2026-06-13T10:30:00.000Z"

scheduledFor

string<date-time>

Accepted initial dispatch deadline. Independent of later continuation wake times.

Example:

"2026-06-13T10:30:00.000Z"

lastHeartbeatAt

string<date-time>

Last heartbeat timestamp.

Example:

"2026-06-13T10:20:00.000Z"

checkpointSummary

string

Bounded terminal or checkpoint summary.

Example:

"Run completed and wrote the release summary."

result

object

Terminal result payload.

Example:

{ "summary": "Deployment looks healthy." }

error

object

Terminal error payload.

Example:

{ "message": "Run failed." }

Last modified on October 5, 2026

Queue an operator response to the latest Agent conversation action

curl --request POST \
  --url https://api.revoengine.com/api/v1/agents/runs/{agentRunId}/actions/{actionId}/resolve \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{}'

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({})
};

fetch('https://api.revoengine.com/api/v1/agents/runs/{agentRunId}/actions/{actionId}/resolve', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

import requests

url = "https://api.revoengine.com/api/v1/agents/runs/{agentRunId}/actions/{actionId}/resolve"

payload = {}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

{
  "agentRunId": "aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa",
  "agentId": "99999999-9999-4999-8999-999999999999",
  "agentSessionId": "abababab-abab-4bab-8bab-abababababab",
  "status": "PROCESSING",
  "serviceAccountUserId": "11111111-1111-4111-8111-111111111111",
  "tickSeq": 3,
  "tickCount": 2,
  "childCount": 1,
  "state": {
    "instruction": "Review deployment"
  },
  "createdAt": "2026-06-13T10:00:00.000Z",
  "lastUpdateAt": "2026-06-13T10:15:00.000Z",
  "version": 1,
  "previousRunId": "aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa",
  "rootRunId": "aaaaaaaa-aaaa-4aaa-8aaa-aaaaaaaaaaaa",
  "parentRunId": "bbbbbbbb-bbbb-4bbb-8bbb-bbbbbbbbbbbb",
  "waitReason": "CHILDREN",
  "triggerType": "REQUEST",
  "triggerRefType": "scheduleId",
  "triggerRefId": "schedule-123",
  "ownerId": "22222222-2222-4222-8222-222222222222",
  "ownerType": "USER",
  "rootAssistantThreadId": "cccccccc-cccc-4ccc-8ccc-cccccccccccc",
  "currentAssistantThreadId": "dddddddd-dddd-4ddd-8ddd-dddddddddddd",
  "lastAssistantMessageId": "eeeeeeee-eeee-4eee-8eee-eeeeeeeeeeee",
  "nextWakeAt": "2026-06-13T10:30:00.000Z",
  "scheduledFor": "2026-06-13T10:30:00.000Z",
  "lastHeartbeatAt": "2026-06-13T10:20:00.000Z",
  "checkpointSummary": "Run completed and wrote the release summary.",
  "result": {
    "summary": "Deployment looks healthy."
  },
  "error": {
    "message": "Run failed."
  }
}

Before you start

Platform API

Queue an operator response to the latest Agent conversation action

Authorizations

Path Parameters

Body

Response