Theme
AI Resources
TaxHacker
TaxHacker is a self-hosted AI accounting app for turning receipts, invoices, PDFs, and transaction records into structured rows that a person can review.
The official repository presents TaxHacker for freelancers, indie hackers, and small businesses that want AI-assisted expense and income tracking. It describes receipt and invoice uploads, extraction of merchant, date, tax, item, and amount fields, custom AI prompts and fields, project and category filters, OpenAI, Google Gemini, Mistral, and local OpenAI-compatible LLM options, Docker/PostgreSQL self-hosting, CSV exports, and an early-development warning. Use this as a first read, not a recommendation. Open the original project before trusting details like terms, limits, privacy, cost, setup, or safety.
What it is
AI extraction for financial documents
TaxHacker is built around uploading receipts, invoices, PDFs, and photos, then using an LLM workflow to extract transaction details into a structured database.
Why it stands out
Custom fields, prompts, and model choices
The useful angle is not just OCR. The README describes custom fields with custom AI prompts, categories, projects, currency conversion including crypto, and a choice between hosted providers and local OpenAI-compatible LLM endpoints.
Availability
Repo, website, releases, and self-hosting path
Readers can inspect the public repository, project website, release page, Docker image notes, Docker Compose path, PostgreSQL requirement, import/export behavior, and setup settings before testing it.
Why it matters
Receipts become reviewable records
TaxHacker is practical when receipts and invoices are scattered across photos, PDFs, currencies, and repeated business rules. The workflow keeps extraction, review, filters, and exports in one place, so a human can check the records before sharing them with an accountant.
How it fits
A workflow app, not just a model
TaxHacker belongs in the useful-tools layer rather than the model layer. The LLM is part of a larger workflow for uploads, field extraction, review, filtering, and exports.
Reporting note
What the source materials list
The README lists receipt, invoice, PDF, and photo handling; merchant, date, tax, item, and amount extraction; custom prompts and fields; categories and projects; currency conversion including crypto; CSV and archive exports; hosted and local LLM options; and Docker/PostgreSQL self-hosting.
Before using
What readers may want to review
The repository warning that the project is still in early development, plus the current release notes and open issues before relying on it for real records.
How financial documents, API keys, uploaded files, model-provider calls, local LLM endpoints, PostgreSQL data, backups, and exports will be handled in the intended setup.
Where extracted merchants, dates, tax fields, item splits, categories, currency conversions, and custom fields need correction after human review.
Local rules, accountant expectations, and record-retention needs before treating any exported report as tax-ready.
The Docker, PostgreSQL, Ghostscript, GraphicsMagick, signup, storage, and network settings before exposing a self-hosted instance.
Reader fit
Who may find it relevant
Freelancers, solo builders, and small teams comparing AI-assisted ways to organize receipts, invoices, and transaction evidence.
Readers who want a self-hosted app where financial-document extraction, review, categories, projects, and exports stay visible.
Builders testing local or provider-based LLM extraction on real document workflows instead of generic chat prompts.
Less relevant for readers looking for a no-setup consumer app or a replacement for professional accounting review.
Editorial note
Why it is included here
TaxHacker shows AI document extraction inside a concrete receipt-and-invoice workflow, with custom fields, categories, projects, currency handling, and exports around the model step.
Source links
Original materials
Reader note
Before relying on this entry
LifeHubber lists entries to help readers inspect AI projects, not to endorse them or prove they are safe, suitable, accurate, maintained, or right for a specific use. We do not verify every entry in depth. Before relying on anything listed, review the original materials, terms, privacy practices, limits, and risks that matter for your situation.
Get occasional updates when new AI resources are added
Occasional notes when new AI resources are added. The form below is handled by the mailing-list service, so its own terms apply when you subscribe.
More in Productivity
Keep browsing this category
A few more places to continue in productivity.
HTML Anything
nexu-io/html-anything
A local agentic HTML editor that uses existing coding-agent CLI sessions to turn Markdown, data, and notes into exportable HTML, PNGs, decks, social cards, data reports, and web prototypes with skill templates and sandboxed preview.
MiMo Code
XiaomiMiMo/MiMo-Code
A Xiaomi MiMo terminal-native AI coding assistant built on OpenCode, with public source, docs, install paths, build/plan/compose agents, persistent memory, task tracking, subagents, provider configuration, and release assets.
Open Design
nexu-io/open-design
A local-first AI design workspace that connects coding-agent CLIs to prototypes, decks, media outputs, design systems, sandboxed previews, and export workflows.
Related in LifeHubber
Keep the thread going
Follow the next layer with AI Resources for AI projects with original links and practical caveats, AI Guides for decision habits for messy AI choices, AI Access for free and low-cost ways to compare AI model access, AI Ballot for a clearer view of what readers are leaning toward, and AI Radar for AI stories that deserve a second look.