TaxHacker is a self-hosted AI-powered accounting application designed for freelancers and small businesses. It automates the processing of receipts, invoices, and transactions by extracting key data using Large Language Models (LLMs), enabling efficient expense and income tracking, categorization, and reporting.

How It Works

The application leverages GenAI to analyze uploaded documents (photos, PDFs) and extract critical information such as vendor, date, amount, and VAT. It supports multi-currency conversion, including cryptocurrencies, using historical exchange rates. Users can customize LLM prompts for specific data extraction needs, create custom fields, projects, and categories to tailor the system to their unique workflows, and filter/export data in CSV format.

Quick Start & Requirements

• Install/Run: curl -O https://raw.githubusercontent.com/vas3k/TaxHacker/main/docker-compose.yml && docker compose up
• Prerequisites: Docker, PostgreSQL 17+ (recommended). For local development: Node.js (v15+), Ghostscript, and GraphicsMagick.
• Setup: Docker Compose setup is straightforward. Local development requires cloning, dependency installation, environment variable configuration, and database initialization.
• Docs: GitHub Repository

Highlighted Details

• AI-driven data extraction from various document types (receipts, invoices, checks, handwritten notes).
• Multi-currency support for over 170 fiat currencies and 14 cryptocurrencies with historical rate conversion.
• Highly customizable LLM prompts and custom fields for tailored data capture.
• Self-hosted architecture with Docker Compose for data control.

Maintenance & Community

The project is actively maintained by vas3k. Contributions are welcome via GitHub issues and pull requests. Donation links are provided.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration with closed-source systems.

Limitations & Caveats

The project is explicitly stated to be in a "very early stage" and should be used at the user's own risk.