File
SMF·RELAY-APP
Received on
31.07.2026
Reviewed on
28.09.2026
Exhibits annexed
3
Questions
4

SMF·RELAY-APP

Can a prompt replace Relay.app?

Automation — workflow automation and app integrations

Almost Verdict recorded on 28.09.2026 · Verified on 31.07.2026
Price
$19/moSource: relay.app · Checked on July 31, 2026
Per year
$228
Build time
A week
Category
Automation
Votes
0 votes
YesAlmost (checked)Not yet

Exhibit tracking slip

Exhibit A The prompt
Exhibit B What you lose
Exhibit C Why people still pay: connector breadth and execution reliability
Exhibit Q Questions

Verdict

A workflow that pauses at a specific step and waits for a human to approve, edit, or reject before continuing — Relay.app's actual differentiator over fully automated tools — is a real weekend build once the automation itself is scoped to two connected services. What doesn't survive: Relay's library of pre-built connectors and its polished approval-request UI across mobile notifications.

Exhibit B — What you lose

Exhibit A — The prompt

Received on31.07.2026
Build a two-service automation with a human-approval step in the middle — Relay.app's actual differentiator over fully automated tools like this catalogue's Pipedream and Gumloop entries. Use TypeScript/Node.js with Postgres. Implement a fixed workflow: trigger (webhook), a data-fetch or transform step, a 'pending approval' step that pauses the run and emails a reviewer a link with the current data and approve/edit/reject actions, and a final action step, calling the second connected service's API, that only runs after approval. Store each run's state explicitly (pending, approved, rejected, completed) in Postgres so a paused run survives a server restart. The approval link should work without requiring the reviewer to log in first — a signed, single-use token is enough for a personal tool. Let the reviewer edit the data before approving, not just accept or reject it as-is. Do not build a library of pre-built connectors for many services, mobile push notifications, or team-shared libraries — those are out of scope; this automates exactly one pipeline between two named services, with one human approval point. Needs a domain for webhook and approval-link handling, a transactional email provider for approval notifications, and an OAuth app registration for whichever connected service requires it.

Opening prefills the prompt — press enter to run it.

Exhibit B — What you lose

  • B.1 a library of pre-built connectors for many services
  • B.2 polished mobile push notifications for pending approvals
  • B.3 team-shared workflow libraries
  • B.4 high-scale durable execution

Prior art

Exhibit C — Why people still pay: connector breadth and execution reliability

A pause-and-wait step is a state machine, buildable in a weekend; a library of pre-built connectors plus a genuinely pleasant approval experience across devices is the much larger ongoing product.

Questions

What happens while a run is waiting for approval?

It's genuinely paused — its state is stored in Postgres, so it survives a server restart, and picks up exactly where it left off once the reviewer responds.

Does the reviewer need an account to approve something?

No — the approval link uses a signed, single-use token, so a personal reviewer can act on it without logging in first.

Can the reviewer change the data before approving?

Yes — editing before approval is a deliberate part of the design, not just a binary accept-or-reject.

What does it cost to run?

Hosting, a domain, and a transactional email provider's free tier for approval notifications — typically a few dollars a month.

Receipt

Already built this yourself?