Skip to main content
POST
Approve every ticked task

Authorizations

X-API-Key
string
header
required

Query Parameters

dry_run
boolean
default:false

Answer what the decision would do, exactly as the real call answers it, and change nothing.

Body

application/json

The ticked rows, and which of them the person had opened.

task_ids
string<uuid>[]
required
Required array length: 1 - 200 elements
opened_task_ids
string<uuid>[]

The ticked tasks the person opened and read before approving. Words the agent wrote on those that nobody changed are recorded as approved as written: a record of their Looks good, which saves no learning.

Maximum array length: 200

Response

Successful Response

What the batch actually did, so the page can say so in one sentence.

skipped
integer
required

How many had already moved on (settled, decided by somebody else, or a task nobody approves), and how many LinkedIn steps LinkedIn could not take now (the sender is not connected), which still wait. Not an error: a selection goes stale while it is read.

updated
integer
required

How many rows the decision applied to.

saved_learnings
integer
default:0

For an approval, how many learnings a Needs work proposed on the approved rows it saved; each writes again the waiting messages it reaches. On a dry run, how many the ticked rows carry, which approving them would save. 0 for a rejection.

saved_learnings_for_all_sequences
integer
default:0

Of saved_learnings, how many apply to every sequence, not only their own.

saved_needs_work
integer
default:0

How many Needs work the approval saved (or would, on a dry run); one Needs work can change several learnings, which saved_learnings counts.

schedule
OutreachApprovalScheduleReadSchema · object | null

For an approval, when the approved work is expected to run: per sender for LinkedIn, and the emails' count and start. Null for a rejection, and when the schedule could not be read (the approval stands either way).