FINANCE

Approver Verdict Router: Approve to Books or Reject to Refund

Listens for an approver's decision on a flagged expense via webhook, then routes approved items to a Postgres approved-spend table and rejected items to a Stripe refund…

CategoryFinance
Enginesim
Difficultyintermediate
Triggerwebhook
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerWebhook with approver verdictHTTP webhook
  • LogicBranch on approved vs rejected
  • ActionRecord approved charge in PostgresPostgreSQLPostgres
  • ActionRefund rejected charge via StripeStripeStripe
  • OutputNotify employee of outcome in SlackSlack

What it does

This workflow acts on the approver's verdict once a flagged expense has been reviewed. A webhook from your review queue carries the charge ID and the decision. Approved charges are recorded in the Postgres approved-spend table with the approver's name and timestamp, closing the loop for the books. Rejected charges trigger a Stripe refund of the disputed amount and a Slack message to the employee explaining which policy was broken and that the charge is being reversed.

When to use it

Use it as the action half of an expense review pipeline — pair it with a queue that surfaces violations. It removes the manual step of someone copying a verdict into the ledger and separately issuing refunds.

How it works

  1. 1A webhook delivers the approver's verdict and charge ID.
  2. 2A branch splits on approved versus rejected.
  3. 3Approved charges are written to the Postgres approved-spend table.
  4. 4Rejected charges are refunded through Stripe.
  5. 5The employee is notified in Slack with the outcome and reason.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect HTTP webhookTrigger any URL on agent actions.
  2. 2
    Connect PostgresAny Postgres URL — query, write, migrate.
  3. 3
    Connect StripeCustomers, subscriptions, payments.
  4. 4
    Connect SlackChannels, DMs, threads, mentions.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.