A local-first personal finance dashboard that pulls your banks, cards, PayPal and subscriptions into a single calm view. No telemetry, no cloud sync, no remote account.

Money that lives in one bank is easy. Money that lives across two banks, a credit card, PayPal and a fistful of app-store subscriptions is not — and the only tool most people have for it is a spreadsheet rebuilt every month.

The usual fix is to hand a third party read access to every account you own. beatrax takes the other route: it reads the statement files your bank already exports, on your machine, and never phones home.

$ # everything beatrax knows lives here, and only here
$ ls ~/Library/Application\ Support/beatrax/
database.sqlite

• Reads the formats European banks already export — CAMT.053 (ISO 20022), MT940 and CSV — so it is not tied to one institution. Plus ICS credit-card PDF statements and PayPal CSV.
• Optional email-receipt scanning via Gmail or Microsoft Graph, so an app-store charge on your card gets matched to the receipt that explains it.
• Optional open banking, off by default, through the Enable Banking aggregator on your own key. Your machine talks to the aggregator directly — there is no beatrax server in the path, because there is no beatrax server.
• Resolves the routing chains between your accounts — PayPal → bank, card → bank via bulk SEPA settlement — so a charge points at whatever really paid for it.
• Sees what is coming — recurring series, drift alerts when a subscription quietly goes up, and a 30/60/90-day cash-flow forecast across every account.
• Budgets in envelopes — assign every euro, roll balances over, move money between categories, and import your history from YNAB, nYNAB or Actual.

beatrax runs entirely on your own machine. The SQLite database, the OAuth tokens, the cached email receipts — all on your disk, and none of it leaves unless you export it yourself. There is no telemetry, no analytics, no cloud account, and nothing to sign up for.

Sync between your own devices is peer-to-peer and end-to-end encrypted: devices pair by QR or word-code with a human-verifiable safety number, talk directly over the LAN when both are awake, and fall back to a store-and-forward relay that only ever holds ciphertext. At rest, the database is encrypted with a passphrase-derived key gated behind the app lock.

One person, or a two-person household, whose spending is spread across several banks, cards and payment processors, and who has given up reconciling it by hand. It assumes you can install a desktop application and grant an OAuth permission if you want receipt scanning.

If you bank with one institution that already gives you a good app, you probably don't need beatrax.

v1.3 shipped. The current development line is v2.0 — a large release, which is why it is v2.0 and not a point bump:

• Local-first end-to-end-encrypted peer-to-peer device sync
• A notification inbox with proactive, locally-generated reminders
• The optional open-banking connector
• Envelope (zero-based) budgeting, with split transactions
• A general-purpose rules engine that does more than assign categories
• Migration importers for YNAB, nYNAB and Actual Budget

Still in flight for v2.0: the mobile client as a fully synced peer, and app-store distribution.

beatrax is licensed under the Hippocratic License 3.0. It is source-available, not OSI-approved open source: read it, run it, modify it — but not as an instrument for human-rights abuse. Software that ingests a person's entire financial history should be auditable, and should say out loud what it won't be used for. The reasoning is written up in NOTICE.md.

If you need an OSI-approved licence for procurement or downstream relicensing, beatrax is not the right project for you.

You don't need to write code:

• Try it and tell us what broke. Clear bug reports are worth more than most patches.
• Improve the docs. Something unclear is a bug; a docs PR is a real contribution.
• Translate. The interface and README already ship in English and Dutch; more are welcome.
• Design & accessibility feedback on any surface in the app.

Prefer to write code? The most useful things are new ingestion adapters (especially Dutch-market sources we haven't covered), categorization rules and merchant heuristics, and email-receipt parsers. Start with CONTRIBUTING.md — it covers the module boundary and the quality gate every PR has to pass.

Not sure where something belongs? File it anywhere; routing is our job — or just ask on Discord.