Hugo D.
Categorizes every transaction. Updates runway daily. Flags anomalies before they bleed.
Hugo keeps the books clean. Pulls transactions, categorizes anything above 0.85 confidence, holds the rest for human review. Recomputes runway daily. Flags spend anomalies and unfamiliar vendors.
What Hugo shipped
Hugo produces real artifacts. Not summaries.
Every run ends in a real deliverable — an email you can send, a doc you can review, a flagged record you can action — not a paragraph of "I ran a tool, here's what I think happened".
Runway · finance/runway.md ready
Cash
$1.8M
Burn
$142k−5%
Runway
12.7 mo
AWS spend+62%· new GPU cluster
Helix LLC$4.2k· new vendor · no PO
184 txcategorized · books up-to-date
Skills
How Hugo works.
Books up-to-date, daily
Every transaction categorized. Confidence < 0.85 flagged. Runway recomputed.
Anomaly radar
Spend > 1.5× trailing average · new vendor without PO · duplicate-looking charges — surfaced.
Runway honest
Cash + burn + AR + AP + signed pipeline. The number you actually have, not the optimistic number.
Slack ping the ops lead
New-vendor without PO at $4k+ pings ops lead before the AP cycle runs.
Tools Hugo uses
Native, no Zapier in the middle. Tool calls land in the audit log.
MercuryQuickBooksBrexSlackVault
One real run
Watch Hugo work.
Hugo D.
Bookkeeper · Finance & Ops
Categorize this week’s transactions and update runway.
done · 8.4s · $0.11
38 categorized · 4 held for review · 2 anomalies surfaced. Runway updated to 12.7 mo. Slack ping sent on Helix vendor.
audit trail · 4 steps[show]
→ ledger.fetch · last_7d
→ classify · 42
→ vault.write · finance/runway.md
→ anomaly_scan
Hugo · today
Tx categorized today
184
Runway
12.7 mo
Anomalies flagged
2+1
Same team
Hugo works with the rest of Finance & Ops.
Hugo is spinning up. Get on the list.
Early access opens shortly. We onboard Finance & Ops customers in cohorts so the bar stays high.