Niches · Human approval & handoff
Reach the person wherever they are and get the answer in time
No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.
What counts as done
The request arrives on at least one of the person's channels, carries enough of the payload to decide or a link to it, can be answered from that channel, and a delivery failure is reported to the agent.
Measured, for our own tests: it arrives within 30 seconds. No docs we read state these figures, so they don't count against a listing here.
How agents ask for it: ping me on my phone when the agent needs me
, send the approval to Slack
, notify me on Telegram
Does the job, with a limit 5
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 |
|---|---|
| gotoHuman D gotoHuman UG (haftungsbeschränkt) | Web inbox, email and a Slack channel post per request, each with a link; external reviewers get a unique link with no account. Answers are given on the review page, not in Slack, and no notification latency or delivery-failure report is documented. source |
| Orkes Conductor Human tasks C Orkes, Inc. | Assignees can be Conductor or external users and groups, but there is no built-in Slack or email prompt: a trigger policy (PENDING, ASSIGNED) starts a second workflow that notifies. Answers go through the UI or API, and no delivery latency or failure report is documented. source |
| Permit MCP Gateway C Permit Inc. | Email and Slack incoming-webhook notifications carry context and a dashboard link (Slack omits tool arguments); decisions are made only in the admin dashboard, approvers must be gateway admins, and no delivery failure is reported to the agent. source |
| Pushary C RalphNex OÜ | Phone push with lock-screen approve or deny, Mac notch, Slack buttons and modals, and a browser page, each answerable in place; noDevices and sent: 0 tell the agent nothing was reached. The body is cut at 180 characters and no delivery time is stated beyond 'instantly'. source |
| withHuman BB Metoro Inc | Email, SMS, WhatsApp, Slack, Discord, phone app and Mac app per reviewer rules, with immediate phone pushes. Slack messages omit arguments and cannot decide (only the apps can), and a failed channel falls back to email without telling the agent; no delivery time is stated. source |
Doesn't do this job 6
Listings in the category that do something else, with the reason.
| Listing | Why |
|---|---|
| Big Arrow BB Franz Enzenhofer (individual, Vienna) | Answers come only from the local Mac screen (an arrow or floating card, optionally spoken). No chat, email or phone channel, which is the map's named blocker for this listing. source |
| Hatchet B Hatchet Technologies, Inc. | No approver inbox, chat, email or phone channel; answers arrive only as events pushed by SDK or incoming webhook. source |
| Inngest BB Inngest Inc | No reviewer inbox, Slack app or email; the guide's Slack message is the builder's own sendSlackMessage code. source |
| Restate BB Restate GmbH | No inbox, Slack app or email; the guide says the agent sends the callback ID to Slack, email or a dashboard itself. source |
| Temporal BB Temporal Technologies Inc. | No inbox, notifications or routing; the page says the pattern needs a separate approval interface that sends the Signal. source |
| Trigger.dev B API Hero Ltd | No reviewer UI, notifications or routing; the builder sends the token or its public access token to their own app. source |
What stops agents finishing
- no chat, email or phone channel (durable-execution listings, bigarrow)
- message over the channel's size limit dropped silently
- approver must be a product user or admin seat
Around it
Before: human-in-the-loop.approve-action. After: notifications.push.
Where this comes from
Weight 19 of 100 in Human approval & handoff: 7 of 26 accounts (approve from phone, Slack and Discord routing, Telegram, missed prompts, length limits)
The same as data: /api/v1/use-cases/human-in-the-loop.reach-approver.json, or the MCP tool get_use_case with id human-in-the-loop.reach-approver. 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.