Personal finance with SQL

Ask detailed questions about your spending and build reports around the answers. Dollar Diary’s query engine combines transactions, tags, discretion levels, items, and custom fields, so you can explore questions beyond the built-in dashboard.

Write your own read-only SQL or describe what you want in plain English and let AI generate it. The editor runs DuckDB SQL and includes a schema browser, saved queries, and results you can copy or download.

Because Dollar Diary hosts the query engine, you do not need to set up a database or export your data first. Connect your AI agent through the MCP server or REST API, then ask your questions in chat and let it run read-only queries for you.

Ask questions your dashboard does not cover

Your tags and annotations give the engine more than merchant names and amounts to work with. For example:

• “Compare my grocery spending month by month this year.”
• “How much did I spend on Family versus Myself last quarter?”
• “Show Dining Out transactions I marked Optional; regret it.”
• “Break down Christmas gifts by the person they were for.”
• “Which merchants account for most of my spending on vacation?”

Use your own tag names, dates, and discretion levels to define exactly what belongs in a report. Explore the power of tags to see how the same purchases can answer different questions.

Generate and modify queries with AI

Start with a question such as “Show my monthly grocery spending this year.” AI generates SQL you can inspect, edit, and run.

Then refine the query with a follow-up instruction:

• “Only include spending for Family.”
• “Compare it with the same months last year.”
• “Break it down by merchant instead.”
• “Show the individual transactions instead of monthly totals.”

You can use AI to modify a query you wrote yourself too. Keep refining the same report as your question becomes more specific. Check that the dates, tags, and discretion levels match what you meant.

See the transactions behind your question

A total tells you how much; the matching transactions show what it includes. For queries that return matching transactions, view them in an inline grid on the Queries page and open individual transactions to inspect their details.

For example, ask for Dining Out transactions marked Optional; regret it, then look through the matching meals and their tags. Switch the question to Family spending to see purchases across groceries, gifts, and other categories.

Save, copy, and download your results

Save a useful query so you can run it again as new transactions arrive. Give it a name that describes the question, such as “Monthly groceries” or “Family spending.”

Copy results into a spreadsheet, or download them as CSV, Excel, or JSON. Copying supports result sets of up to 500 rows; use a download for larger results.

Write SQL when you want direct control

Use the schema browser to explore the available tables and fields, then write or edit SQL directly with syntax highlighting. Combine filters, group spending, compare periods, and calculate your own totals without fitting the question into a preset report.

Queries are read-only. Run them to analyze your workspace’s spending; edit transactions through the app.