Bookbot

Finds the right accounts for your transactions. Bookbot looks at how you recorded similar transactions before and suggests the missing From or To account, and the custom properties they always share.

Apps, scripts, and agents ask Bookbot for accounts through its API.

Bookbot never posts transactions and never replaces an account or property you already set.

How it works

Account discovery

Bookbot assigns missing From and To accounts by searching your book's posted transactions for similar patterns:

1. Identify key terms — AI picks out meaningful parts of the description (e.g. Office Depot from OFFICE DEPOT #1234 PURCHASE 04/15)
2. Search by account + creator — finds posted transactions with similar terms, scoped to the known account and user
3. Search by account — same search, any user
4. Search by description — searches across all accounts
5. Pick by frequency — assigns the most commonly used accounts from matching results

Accounts already on the transaction are kept. Bookbot only fills the sides that are missing.

When every matching transaction shares the same custom property (for example, a consistent category or project code), Bookbot copies that property onto the draft. It only fills missing properties and never overwrites properties already present. You can prevent selected property keys from being copied with agent_excluded_properties.

Re-checking drafts after a correction

When you correct an account that Bookbot suggested on a posted transaction, Bookbot re-runs discovery on the pending drafts in the book that received the same suggestion — one correction can fix many. The re-check considers up to 100 pending drafts and is skipped when the book's backlog has more than 50 items.

Customize description matching

Use agent_discovery_examples to show Bookbot which recurring parts of a transaction description are useful for finding similar posted transactions. It improves the search terms; it does not assign an account directly.

Add one example per line as full description -> [stable part]:

agent_discovery_examples: |-
  WF HOME MTG DES:AUTO PAY ID:XXXXX37704 CO ID:XXXXX87461 WEB -> WF HOME MTG DES
  AMAZON RETA* NJ1JM7QA0 #goodies -> AMAZON RETA
  1* Product Alpha * 2* Product Beta -> [Product Alpha, Product Beta]

• On the left, use a real description from your transactions.
• On the right, keep recurring names or account identifiers; omit changing dates, amounts, quantities, and reference IDs.
• Use brackets with comma-separated values when more than one part is useful.

Set the property on an account for account-specific formats, an account group for shared formats, or the book as a general fallback. Bookbot uses the first non-empty value in account → first matching group → book order; values are not combined. When no account is known yet, only the book value applies. Custom examples replace the built-in examples.

AI allowance behavior

Discovery uses Bkper AI to pick out the key terms of a description, with Gemini Flash Lite and minimal reasoning. Inference runs with the allowance of the signed-in user, or of the installation principal for event-driven re-checks, and is attributed to Bookbot. For Business and Professional subscriptions, the Bkper AI allowance may be shared across the domain. The Bkper AI usage dashboard shows the usage attributed to Bookbot.

Configuration

API access

Production: https://bookbot.bkper.app
Preview:    https://bookbot-preview.bkper.app
OpenAPI:    https://bookbot.bkper.app/openapi.json

# Run `bkper auth login` first if needed
TOKEN="$(bkper auth token)"

curl -X POST \
  -H "Authorization: Bearer ${TOKEN}" \
  -H "Content-Type: application/json" \
  -d '{
    "itemId": "1",
    "bookId": "<book-id>",
    "transaction": { "description": "Monthly rent", "creditAccount": { "id": "<account-id>" } }
  }' \
  https://bookbot.bkper.app/api/v1/discovery

Learn more

• Recording transactions — how to record and manage transactions in Bkper