INVOICE PROCESSING

Screen incoming invoices for duplicates and fraud signals

On each emailed invoice, extracts vendor, invoice number, and bank details, checks Postgres for prior payments and changed payee accounts.

CategoryInvoice Processing
Enginesim
Difficultyintermediate
Triggerevent
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerInvoice received in inboxGmailGmail
  • ActionOCR vendor, number, amount, bank detailsOpenAI
  • ActionLook up prior payments and verified bank accountPostgreSQLPostgres
  • LogicDetect duplicate number or changed payee account
  • ActionQuarantine suspect invoicePostgreSQLPostgres
  • OutputAlert finance-controls channelSlack

What it does

Every inbound invoice is screened before it ever reaches approval. The flow OCRs the document, then checks two risks: a duplicate (same vendor and invoice number already paid) and payment fraud (the vendor's bank details differ from the last known account on file). Either condition quarantines the invoice and raises an alert.

When to use it

Use this as a front-line control when you process high volumes of vendor invoices and worry about double-payments or business-email-compromise scams that swap in attacker bank accounts.

How it works

  1. 1A vendor invoice arrives in the inbox and triggers the run.
  2. 2OCR pulls vendor name, invoice number, amount, and bank/routing details.
  3. 3Postgres is queried for any prior invoice with the same vendor and number, plus the vendor's last verified bank account.
  4. 4A risk check flags exact-duplicate numbers and any mismatch in payee bank details.
  5. 5Suspect invoices are written as quarantined so they cannot be paid.
  6. 6A detailed fraud/duplicate alert is sent to the AP and finance-controls channel.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect GmailRead, draft, send, label.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect PostgresAny Postgres URL — query, write, migrate.
  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.