Budget Category Suggester from Statement

WHAT IT DOES

AI-powered budget category suggester that analyzes bank or credit card statement transactions and suggests appropriate budget categories for each purchase. The script reads transaction details from the current page—merchant names, amounts, and dates—and generates a categorized list with suggested budget categories like Groceries, Dining, Transportation, Utilities, Entertainment, Shopping, Housing, Insurance, Healthcare, and Education. You can customize the output format (list/table/compact), category system (simple/detailed/custom), and toggle inclusion of confidence levels, amounts, and dates. The user reviews the output before copying or using it.

WHERE IT RUNS

Works on all bank and credit card statement pages (match pattern *://*/*), including online banking portals, credit card statement pages, expense tracking platforms, financial dashboards, and any page containing transaction data.

HOW TO USE

Install the script and the Budget Category Suggester appears in the panel. Configure your preferred output format, category system, and toggle inclusion of confidence levels, amounts, and dates using the settings controls. Click "Suggest" to analyze the current page. The AI extracts transaction data and suggests budget categories for each item. Review the output in the panel. Click "Copy to Clipboard" to copy the categorized list for use in budgeting apps, spreadsheets, or financial planning. Adjust the settings and regenerate if you want a different format or category system.

REAL MECHANISM: DOM CONTENT EXTRACTION WITH AI CATEGORY SUGGESTION AND REVIEW GATE

This script uses a hybrid extraction approach that prioritizes structured content when available. First, it checks for WebMCP (navigator.modelContext), a W3C proposal that exposes structured tools on the page. If available, it calls getContent() to obtain structured transaction data. If WebMCP is not present—the common case today—it falls back to DOM scraping using targeted selectors for statement sections (transaction-table, statement-table, transactions, transaction-list, account-activity, and transaction-details). It also scans table structures for transaction keywords (merchant, vendor, transaction, date, amount, payment, withdrawal, deposit). The extracted text is sent to an AI backend via GM_xmlhttpRequest, which processes it with a system prompt that includes the user's category system preference, output format, and inclusion preferences. The AI returns a categorized list (up to 300 words) with suggested budget categories for each transaction, formatted according to user preferences. The response is displayed in a bordered container for review before copying. Copy-to-clipboard uses GM_setClipboard. All settings (outputFormat, categorySystem, includeConfidence, includeAmounts, includeDates) are persisted via GM_setValue. The script enforces a review-before-use pattern: the user must read and approve the AI output before copying it.

THE PANEL

Drag the panel anywhere on screen by clicking and holding the header. It snaps to screen edges when released. Collapse it to a small bubble by clicking the minimize button, and expand it back by clicking the bubble. Resize the panel from any corner. Press Escape to close the configuration panel. Your position and size preferences are saved automatically per site.

PLEASE NOTE

This script requires the free BotGentz browser extension and an active AI subscription for the suggestion service. AI-generated categories should be reviewed for accuracy before use—the AI may misinterpret merchant names or suggest categories that don't match your personal budget structure. The script reads the current page content and sends it to the AI backend for processing; no data is stored or logged. Websites change frequently; if the script stops detecting transaction content, check for updates to the extraction selectors.