Can Auritrack be vibe coded?
AI-native expense tracker that does your bookkeeping from chat messages and bank statements
The core loop, logging expenses by typing a sentence into a chat box and having an LLM extract amount, payee, and category, is very buildable with an Anthropic API key. Budgets, categories, and spending reports are standard CRUD. The real gap is statement import: getting one clean CSV parsed by an LLM is a demo, but reliably parsing messy multi-page bank PDFs across many banks needs batching, fallbacks, and cross-checks that take far longer than a contained effort.
Jump to the build brief ↓Checked Aug 2026
What you pay today, before any DIY hosting
high editorial confidence
Buildability by layer
Screens, forms, and focused interactions
The repeatable job the product performs
Availability and legality of required data
Uptime, queues, support, and maintenance
Security, compliance, and user confidence
The achievable core
- Build a chat input that sends free-text expenses to an LLM with a strict extraction schema, save transactions to SQLite, auto-create categories, track budgets per month, and chart spending.
- Build a focused single-user workflow with real persistence, search, and export.
- A responsive interface with real empty, loading, success, and error states.
The parts a prompt cannot buy
- battle-tested statement parsing across many banks and layouts
- mobile apps and push notifications
- Telegram bot logging
- predictive spending forecasts
- The last 20 percent is sync, migration fidelity, speed, and edge cases.
Why people still pay
They pay because a hardened parsing pipeline, mobile apps, and chat-anywhere logging remove all friction, and $3/mo is cheaper than maintaining your own LLM plumbing.
The last 20 percent is sync, migration fidelity, speed, and edge cases.
The brief
Context, requirements, acceptance criteria, non-goals, and the full production standard — as Markdown, ready for any coding agent.
Build brief — a focused alternative to Auritrack
Context
**Auritrack** — AI-native expense tracker that does your bookkeeping from chat messages and bank statements. It currently costs $3/mo.
The core loop, logging expenses by typing a sentence into a chat box and having an LLM extract amount, payee, and category, is very buildable with an Anthropic API key. Budgets, categories, and spending reports are standard CRUD. The real gap is statement import: getting one clean CSV parsed by an LLM is a demo, but reliably parsing messy multi-page bank PDFs across many banks needs batching, fallbacks, and cross-checks that take far longer than a contained effort.
This brief describes a focused, single-operator replacement for the part of Auritrack that is genuinely reproducible. It is deliberately narrower than the product it replaces, and it says so in writing. Build the useful core; do not pretend to have rebuilt the rest.
What you are building
Build a chat input that sends free-text expenses to an LLM with a strict extraction schema, save transactions to SQLite, auto-create categories, track budgets per month, and chart spending.
Build a focused single-user workflow with real persistence, search, and export.
A responsive interface with real empty, loading, success, and error states.
Requirements
Functional
Local or hosted database.
CSV import.
PDF text extraction.
Charting library.
Data and integrations
Anthropic API key.
Each of these needs a real account, credential, or quota. Set them up before writing feature code.
Non-functional
Accessibility: semantic markup, labelled controls, visible focus, and reduced-motion support.
Security: server-side secrets, validated input, and no credentials in the client bundle.
Reliability: retries with backoff on external calls, and a clear failure state when a provider is down.
Portability: the operator can export their data and leave without losing it.
Implementation brief
Build me an AI expense tracker to replace Auritrack. Requirements:
Local web app: Node + Express + better-sqlite3, server-rendered, binds to
localhost only.
Chat box on the home page: I type "lunch 12.50 at Chipotle yesterday" and the
app calls the Claude API (claude-sonnet-5) with a tool schema to extract
amount, payee, category, and date, then saves the transaction. Show the parsed
result inline with an edit button so I can correct mistakes.
Categories are auto-created the first time the model uses one; also give me
plain manual add/edit forms for transactions and categories as a fallback.
Statement import: upload a CSV or a text-layer PDF (use pdf-parse). Send rows
or pages to the model in batches with a strict JSON schema, show everything in
a review table before committing, and dedupe on date + amount + payee.
Budgets: monthly limit per category with a progress bar that turns red on
overspend.
Reports: spending by category per month and a 6-month trend line with
Chart.js, plus an "ask about my spending" box where the model writes a SQL
query, runs it against a read-only connection, and explains the answer.
Nightly copy of the SQLite file to backups/, keep 30.
ANTHROPIC_API_KEY lives in .env; send the model only the rows a request
needs and never log transaction data.
Out of scope: bank sync, scanned-image PDFs, mobile apps, Telegram bots, and
multi-user. Note in the README that imports cost real API money and roughly
how much per statement.
README: how to map my bank's CSV columns on first import.
Delivery standard
Inspect the repository first, then write a short implementation plan before writing code.
Deliver the smallest complete end-to-end workflow first; every primary control must work against persisted data.
Use real validation and storage; never substitute fake dashboards, decorative controls, hard-coded success states, or mock integrations.
Include responsive layouts plus genuine empty, loading, success, validation, and failure states.
Keep secrets server-side in environment variables, provide .env.example, and never commit credentials or user data.
Add structured logs around every external call and return actionable errors without leaking sensitive details.
Write unit tests for the core logic and one automated test of the main user journey.
Finish with a README covering setup, architecture, data location, backups, tests, deployment, and known limitations.
Acceptance criteria
A clean install starts the app using only the README and .env.example.
The primary journey works from first visit through saved result, reload, edit, export, and deletion where applicable.
Invalid input, missing configuration, provider failure, and an empty database each have a usable state.
The interface works at 390px and 1440px, is keyboard navigable, and shows visible focus on every control.
Tests, type checking, linting, and a production build all pass with no ignored failures.
No part of the interface implies a live integration, security guarantee, or scale capability that was not actually built and verified.
Non-goals
Do not build these, and do not claim to have replaced them:
Battle-tested statement parsing across many banks and layouts.
Mobile apps and push notifications.
Telegram bot logging.
Predictive spending forecasts.
The last 20 percent is sync, migration fidelity, speed, and edge cases.
What you still own after launch
Secure credentials, rotate secrets, and handle provider rate limits.
Run migrations, backups, restores, and dependency updates.
Test the critical journey after every model, API, or hosting change.
Monitor failures and fix the edge cases a first prompt will miss.
Risk
**High consequence.** Use this as a prototype or personal aid. Keep a qualified human and an established provider in the loop for consequential decisions.
Editorial confidence in this assessment: high. No independent one-shot implementation is linked yet.
Prior art
Working open-source software you can read, fork, or borrow from before starting:
[Firefly III](https://github.com/firefly-iii/firefly-iii) — Mature self-hosted personal finance manager with a full API and CSV importer, a solid non-AI base to bolt an LLM onto
[Actual Budget](https://github.com/actualbudget/actual) — Open-source local-first budgeting app with strong transaction import tooling
Generated by [Can It Be Vibe Coded?](https://www.canitbevibecoded.com) · Full report: https://www.canitbevibecoded.com/auritrack
You still own the product
- Secure credentials, rotate secrets, and handle provider rate limits.
- Run migrations, backups, restores, and dependency updates.
- Test the critical journey after every model, API, or hosting change.
- Monitor failures and fix the edge cases a first prompt will miss.
Open-source prior art
Before you start
Can Auritrack be vibe coded?
Partly, if you narrow it. The core loop, logging expenses by typing a sentence into a chat box and having an LLM extract amount, payee, and category, is very buildable with an Anthropic API key. Budgets, categories, and spending reports are standard CRUD. The real gap is statement import: getting one clean CSV parsed by an LLM is a demo, but reliably parsing messy multi-page bank PDFs across many banks needs batching, fallbacks, and cross-checks that take far longer than a contained effort.
What can an AI coding agent reproduce from Auritrack?
Build a chat input that sends free-text expenses to an LLM with a strict extraction schema, save transactions to SQLite, auto-create categories, track budgets per month, and chart spending. Build a focused single-user workflow with real persistence, search, and export. A responsive interface with real empty, loading, success, and error states.
What will a DIY Auritrack replacement still be missing?
battle-tested statement parsing across many banks and layouts; mobile apps and push notifications; Telegram bot logging; predictive spending forecasts; The last 20 percent is sync, migration fidelity, speed, and edge cases.
What do I still own after building a Auritrack alternative?
Secure credentials, rotate secrets, and handle provider rate limits. Run migrations, backups, restores, and dependency updates. Test the critical journey after every model, API, or hosting change. Monitor failures and fix the edge cases a first prompt will miss.