INVOICE PROCESSING
Silent Price-Creep Detector from Gmail Invoices
Watches an inbox for incoming vendor invoices, extracts each line item's unit rate, compares it to the last invoice from that vendor.
How it runs
The automated pipeline, trigger to output.
- TriggerNew invoice email arrives in GmailGmail
- ActionExtract vendor and per-SKU unit rates with OpenAIOpenAI
- ActionLook up last stored rate per SKUPostgres
- LogicCompute rate delta and check for prior notice
- ActionUpdate baseline rate in ledgerPostgres
- OutputPost flagged price creep to SlackSlack
What it does
This workflow catches vendors who quietly bump unit prices on recurring invoices. Each time a new invoice lands in Gmail, it parses the line items, looks up the prior unit rate per SKU from your ledger, and flags any increase that arrived with no price-change notice. Confirmed creep is posted to a Slack review channel with the old rate, new rate, and percentage delta.
When to use it
Run this when you pay the same vendors on a recurring basis (SaaS seats, raw materials, logistics) and want to stop rubber-stamping invoices that crept up 3-8% without anyone noticing. Best for finance and procurement teams processing invoices by email.
How it works
- 1A new email matching your invoice label arrives in Gmail and triggers the run.
- 2An OpenAI extraction step pulls vendor, invoice date, and per-SKU unit rates from the attachment or body.
- 3The flow queries Postgres for the most recent stored rate per SKU for that vendor.
- 4A logic step computes the delta and checks whether a notice email exists; small or pre-announced changes are filtered out.
- 5Confirmed silent increases are written back to Postgres as the new baseline.
- 6A Slack message delivers the flagged creep to the procurement channel for sign-off.
Set it up
What you configure once, before turning it on.
- 1Connect GmailRead, draft, send, label.
- 2Connect OpenAIModels, embeddings, files.
- 3Connect PostgresAny Postgres URL — query, write, migrate.
- 4Connect SlackChannels, DMs, threads, mentions.
- 5Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
- 6Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
- 7Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.
More Invoice Processing workflows
Dollar-Threshold Match Variance Escalation Chain
Routes three-way match variances up an approval ladder by dollar size — small ones to the line manager, large ones to the controller.
Front vendor invoice to Airtable approval packet with line-item parse
Watches a Front inbox for vendor invoice emails, extracts header and line-item data from the body and PDF attachment, and writes a structured.
Gate invoice approvals on a duplicate cross-check
When an approver clicks Approve in your AP system, a webhook re-validates the invoice against paid history in Postgres and Stripe charges.
Three-Way Match Exception Router by Variance Type
When a new invoice fails three-way match against its PO and receiving record, this workflow classifies the mismatch (price, quantity, or missing receipt) and routes each variance…
Route parsed Front invoices to the right Slack approver by amount
Parses an incoming Front vendor invoice and posts an interactive Slack approval card to the approver whose spending threshold matches the invoice total.
Three-way match Front invoices against purchase orders in Postgres
Parses a Front vendor invoice and runs a three-way match against the matching purchase order and receipt records in Postgres, posting pass or exception results to Slack.
Run it inside a business
This workflow drops into a full company template. Import the org, and this is one of the playbooks its agents run.
Run this workflow in your colony.
14-day trial. No DevOps. No Sales call. Provisioned in under a minute.