Can I vibecode Finary?

KINDA · weekend project

price $14.5/moyou'd save $174/yrbuild time a weekendcategory 💰 personal financereplaced by 0 people

The dashboard half of Finary is a weekend project: a table of holdings, a price fetcher, a few charts, a history table so you can see net worth over time. The hard half is the plumbing, connecting to hundreds of European banks and brokers through PSD2 aggregators, which needs either a licensed provider or a lot of scraping you do not want to maintain. You can get most of the way with CSV imports from your broker plus public price APIs for equities and crypto, refreshed on a cron. That gets you honest numbers with a monthly chore attached. If the chore is the reason you were paying, this is not a replacement.

the prompt

Build a local personal net worth tracker. Single user, runs on my machine, no accounts, no cloud, no telemetry.

Stack, non negotiable: Next.js 15 with the App Router and TypeScript, Tailwind for styling, SQLite through better-sqlite3 with plain SQL migrations in /db, Recharts for charts. No ORM, no auth library, no external UI kit.

Data model:
- accounts: id, name, institution, type (cash, brokerage, retirement, crypto, property, loan), currency
- holdings: id, account_id, symbol (nullable), quantity, manual_value (nullable), cost_basis (nullable)
- prices: symbol, date, close, currency
- snapshots: date, total_value, breakdown_json

Features:
1. CRUD pages for accounts and holdings. Holdings are either priced by symbol or carry a manual value for things like property and cash.
2. CSV import at /import: paste or upload a CSV, map columns to symbol, quantity, cost basis, pick the target account, preview before commit.
3. A price refresh route that pulls crypto quotes from CoinGecko and equity or ETF quotes from a single configurable HTTP quote endpoint. Cache into the prices table, never hit the network on page render.
4. A snapshot script, npm run snapshot, that recomputes total net worth and writes one row to snapshots. Document how to run it daily with cron or Task Scheduler.
5. Dashboard: total net worth, net worth line chart from snapshots, allocation donut by asset type, table of holdings with value, weight, and gain versus cost basis.
6. Multi currency: store a base currency in .env, fetch FX rates in the same refresh job, convert everything to base for display.

Out of scope, do not attempt: bank or broker account linking, open banking or PSD2 aggregators, credential scraping, mobile apps, multi user support, tax reporting.

Secrets in .env.example: BASE_CURRENCY, COINGECKO_API_KEY, QUOTE_API_URL, QUOTE_API_KEY, FX_API_URL. Never commit real keys.

Seed the database with three fake accounts and eight holdings so the dashboard is not empty on first run. Write a README with setup, refresh, and snapshot instructions in under thirty lines.

$ open in your agent (prompt prefilled, you press enter) or copy it raw · this prompt is generated from the build plan · improve it via PR

why people still pay

Because nobody actually keeps a spreadsheet updated for two years. Finary's pitch is that you connect once and then your net worth is just correct every morning, including the accounts you forget about, and that only works if someone maintains regulated connections to your specific bank in your specific country. The DIY build is honest and private and free, but it asks you for fifteen minutes of CSV wrangling a month, and most people quietly stop doing that around month three.

what you lose

xAutomatic bank and broker sync across hundreds of institutions

xMobile apps with push updates and Face ID

xAutomated real estate valuation estimates and loan tracking

xBenchmarking against other users and the community and content layer

xSomeone else fixing the connector the day your bank changes its API

prior art · use these instead of building, if you'd rather

Nothing worth pointing at. That's why the prompt exists.

share on X ↗"I just replaced Finary ($14.5/mo) with one prompt"