FINANCE

Accrual Agent: Reconcile POs and Draft Booking Recommendations

An agent pulls open POs, receipts, and invoices from Snowflake and the Coda tracker, reconciles which receipts are genuinely unbilled.

CategoryFinance
Enginepaperclip
Difficultyadvanced
Triggerschedule
Steps6
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerClose-cutoff schedule fires
  • ActionQuery POs, receipts, invoices in SnowflakeSnowflakeSnowflake
  • ActionRead already-booked items from CodaCodaCoda
  • LogicAgent reconciles lines, drafts accrual recommendations
  • LogicHold low-confidence lines for review
  • OutputPost summary to Slack for sign-offSlack

What it does

This is the judgment-heavy version of the accrual check. An agent gathers PO, receipt, and invoice data from Snowflake, cross-references the manual accrual tracker in Coda to avoid double-counting items already booked, and reasons about partial receipts, returns, and timing differences. It then drafts a recommended accrual amount and GL account per line and packages everything for a controller to approve.

When to use it

Use it when simple quantity matching produces too many false positives and you want a first-pass reconciliation that handles partial deliveries, credits, and already-tracked items before a human reviews. Best for month-end when the unbilled list is long and messy.

How it works

  1. 1A schedule triggers near the close cutoff.
  2. 2The agent queries Snowflake for POs, receipts, and posted invoices.
  3. 3It reads the Coda tracker to exclude accruals already booked or in progress.
  4. 4The agent reconciles each PO line, accounting for partials and returns, and drafts a recommended accrual and account.
  5. 5A logic step holds low-confidence lines for manual review.
  6. 6A summary with recommendations posts to Slack for controller sign-off.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect SnowflakeWarehouses, queries, shares.
  2. 2
    Connect CodaDocs, packs, automations.
  3. 3
    Connect SlackChannels, DMs, threads, mentions.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    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.