W1: Auto-tagging

Assignment brief — Workflow 1

Agent that codes corporate-card and bill-pay transactions to the right general ledger account, tax code, and tracking categories, syncs to the customer's accounting system, and learns from every accountant override.

Mid-market multi-currency businesses on Reap process ~250 transactions/month. Today, those land in Xero/QuickBooks Online/NetSuite as uncoded entries that a controller or outsourced bookkeeper hand-classifies — typically 30–90 seconds per line for routine items, longer for ambiguous ones. Month-end close stalls on this queue. Silent miscoding is the worst outcome: it corrupts P&L, VAT/GST filings, and management reporting, and is costlier to unwind than a delayed entry.

This is the skill I would show first. It is built end to end, and it classifies card authorizations, stablecoin payouts, and bills into general-ledger codes only after deterministic shortcuts and tenant-scoped evidence have had first say.

Queue surface: the reviewer sees outcome, confidence, evidence, version stamps, and the override path in one register.

Submission proof

37golden auto-tagging cases

4refusal reason families

$0model cost on vendor-rule hits

1override loop into vendor rules

What runs

EVFinanceEventCard, payout, and bill events arrive in one typed shape.

→

RULEVendor memoryKnown vendor + tenant rule emits a deterministic coding.

→

ModelTyped classifierOnly misses call generateObject with chart-of-accounts and vendor evidence.

→

GATEAutonomy policyConfidence maps to apply, suggest, or refuse before logging.

Why I made this the lighthouse skill

Large language model earns its placeClassification has language ambiguity, vendor-tail risk, and tenant-specific chart-of-accounts mapping. I use a model here because it is useful, but only inside constraints.Schema-checked output only.

Rules beat promptsSecond encounter with a corrected vendor can skip the model entirely.Cost drops, reversals drop.

Refusal is visibleOff-chart-of-accounts, missing input, ambiguous match, or unsafe output become structured refusals.No silent miscoding.

Review compoundsOperator override writes the corrected coding and supersedes the vendor shortcut.The reviewer’s work becomes future automation.

Read next

• Deep dive — gates, calibration, learning loop, and production seams
• Evals — method & results — golden-set composition, model sweep, refusal metrics, replay