Skip to main content

Bkper Agent

Your AI agent for fast, automatic transaction categorization for all your transactions from any source.

Updated today

The Bkper Agent automates the tedious parts of bookkeeping by combining two powerful capabilities: intelligent data parsing and smart categorization.

Turn Documents into Transactions Instantly.

The Bkper Agent is always doing the hard work of interpreting your data and correctly categorizing it in transactions for you, whether it comes from Google Sheets, Bkper Bank Connections, or File uploads.

This power is especially evident when uploading documents. Drag and drop an Invoice, Receipt, or bank statement (as an image or PDF) directly into your book, and the Agent performs its two core functions in one seamless action: intelligent parsing and smart categorization. It parses the document to extract essential data like Dates, Amounts, and Descriptions while simultaneously referencing your book history—and learning from your patterns—to apply the correct Accounts to each new transaction.

The result is a list of fully populated and categorized draft transactions ready for your review and post, whether from a single coffee shop receipt or a multi-page corporate bank statement.

Categorization: Finding the Right Accounts

Once transaction information is available—whether from a document or any other source—the Agent helps to complete the transaction by giving the From and To accounts. This happens automatically based on patterns it recognizes from your bookkeeping history.

The Agent follows a logical sequence when searching for the right accounts. It checks each method, described below, in order and stops as soon as it finds a match. Think of it like a decision tree that becomes more specific with each step.

First, it looks for explicit account names in the description.

If you write "Bank Household rent 1900," and both "Bank" and "Household" accounts exist in your book, the Agent assigns them as the source and destination accounts respectively. The first account name becomes the source (where money comes from), and the second becomes the destination (where money goes to).

If no account names are found, it searches for matching descriptions. When you've previously recorded "Bank Household rent 1900" and later enter just "rent 2000," the Agent recognizes this description and applies the same accounts you used before. This saves you from repeatedly specifying accounts for recurring transactions.

Next, it checks for matching hashtags. If you've used a hashtag like #rent in a previous transaction, you can simply enter "#rent 2000" for the next one. The Agent will use the same accounts associated with that hashtag, making categorization even faster.

Finally, for mobile transactions, it considers location. When you record a transaction at a physical place—say, posting "Wallet Food bakery 23" at your local bakery—the next time you're at that location, you only need to enter the amount. The Agent remembers the accounts you used there before.

Every action the Agent takes appears in your activity history, so you can see exactly what changed and when. This creates a transparent record of the automation at work.

Ignoring Unwanted Text

Sometimes transaction descriptions include information you want to keep but don't want the Agent to process—like timestamps from integrations or reference numbers. You can wrap any text in quotes to tell the Agent to ignore it for matching purposes.

For example, if you enter 10 Gas "at 10:56", the Agent will only use "10 Gas" for finding accounts, but the complete description including "at 10:56" will be saved with your transaction. This proves especially useful when integrating systems that automatically append metadata to every transaction.

Did this answer your question?