Niches · Human approval & handoff
Decide what happens when nobody answers
One company's docs show a full fit.
What counts as done
At the deadline the request ends with the configured outcome, which defaults to deny, the agent gets a typed timeout distinct from rejection, and an escalation or reminder is sent when configured.
How agents ask for it: if I don't answer in an hour, cancel
, escalate to my lead after 30 minutes
, don't do it if no one approves
Does the job 1
The docs show it meets the test above, and nothing we read says otherwise.
Does the job, with a limit 8
It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.
| Listing | Why |
|---|---|
| Hatchet B Hatchet Technologies, Inc. | A deadline needs a SleepCondition in an or group with the event condition; the task then checks which fired. No default outcome, no typed timeout and no escalation or reminder. source |
| Inngest BB Inngest Inc | Every wait takes a timeout and resumes with null or undefined, distinct from a rejection event. What a timeout means and any reminder or escalation are the builder's code; there is no default deny. source |
| Orkes Conductor Human tasks C Orkes, Inc. | slaMinutes per assignment escalates down a chain of assignees, TERMINATE fails the workflow and a TIMED_OUT trigger can notify. The default is LEAVE_OPEN and 0 never expires, so an unset deadline does not deny. source |
| Permit MCP Gateway C Permit Inc. | A timeout always rejects (default 5 minutes) and History marks it timed out, but the window is too short for overnight review, extends only by a +5 min button, and the agent's error is not documented as distinct from a rejection. source |
| Pushary C RalphNex OÜ | An expired status is distinct from deny and is documented as no approval, and an unanswered question can trigger a phone call. Expiry follows server policy with no per-request deadline (timeoutMs sets only one call's wait), and the default deny rests on the skill telling the agent so. source |
| Restate BB Restate GmbH | orTimeout throws a TimeoutError distinct from a rejection, and the timer survives restarts. Without it the wait never ends; no default outcome, reminder or escalation exists. source |
| Temporal BB Temporal Technologies Inc. | The wait takes a timeout and returns a timeout path distinct from a rejection, and the page shows escalation and multi-level chains. All of it is builder code; with no timeout the workflow waits indefinitely. source |
| Trigger.dev B API Hero Ltd | createToken takes a timeout and forToken returns ok: false with an error, distinct from a rejected output. Omitting timeout means no timeout at all (the map's named blocker), and there is no default outcome or escalation. source |
Doesn't do this job 2
Listings in the category that do something else, with the reason.
| Listing | Why |
|---|---|
| Big Arrow BB Franz Enzenhofer (individual, Vienna) | Cards have no time limit by design; they stay until answered, closed or stopped. Arrows end after their duration but report no timeout outcome or escalation. source |
| gotoHuman D gotoHuman UG (haftungsbeschränkt) | The create-review reference has no deadline, expiry or timeout field, and fetch-review knows only pending and responded states, so a request never ends by itself. source |
What stops agents finishing
- no timeout at all, so the wait never ends (Trigger.dev token without timeout, Restate without orTimeout)
- default proceeds on best judgement
- window too short for overnight review (Permit 5 minutes)
Around it
Before: human-in-the-loop.reach-approver. After: notifications.push.
Where this comes from
Weight 11 of 100 in Human approval & handoff: 4 of 26 accounts (AskUserQuestion proceeding after 60 seconds, timeout-as-deny in n8n, Inngest timeouts); buyer check 2
The same as data: /api/v1/use-cases/human-in-the-loop.timeout-policy.json, or the MCP tool get_use_case with id human-in-the-loop.timeout-policy. Other jobs in this category: /api/v1/niches?category=human-in-the-loop.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.