TaxHacker
Self-hosted AI accounting app. LLM analyzer for receipts, invoices, transactions with custom prompts and categories
Runssource: GitHubhomepageTypeScriptMITcommit 6cb7254b854b
TypeScript, MIT licensed. The project labels itself: accounting, ai analysis, currency exchange, expenses, gemini, invoices, llm and ocr recognition.
TaxHacker runs. An Argusic agent installed it in 14 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
TaxHacker installs, builds, and launches successfully with local PostgreSQL 16 and Node.js 22; all 33 tests pass.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 96 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- PostgreSQL not installed. Extracted postgresql-16 debs from Ubuntu archive.5 minutes
- Node.js v18 found but project requires >=26. Installed v22.14.0 via prebuilt binary.2 minutes
- Prisma adapter could not find libpq library at runtime.1 minute
- Install time
- 14 minutes
- Cold machine to finish
- 17 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- Model tokens used
- 63,128
- Exact commit tested
- 6cb7254b854b
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- How the result was proved
- runs before 2026-09-07 did not record the command used to check the result
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
**What works now:** - `npm install` — dependencies installed (860 packages) - `npx prisma generate` — Prisma Client generated - `npx prisma migrate deploy` / `npx prisma db push` — database schema applied - `npm test` — all 7 test files, 33 tests passing - `npm run build` — production build succeeds with all routes - `npm run dev` — dev server starts on port 7331, serves pages (redirects to `/self-hosted/redirect… - PostgreSQL 16 is running locally, database `taxhacker` is initialized and migrated **Limitations (non-blocking):** - Ghostscript (`gs`) not installed — PDF uploads with AI analysis won't work at runtime (PDF previe… - Ghostscript can't be installed without root; would need `apt-get install ghostscript` or a Docker… - Node is v22 instead of the requested >=26, but all dependencies support it and everything builds/…
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 14 minutes.
- Recovered from all 3 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
- Measured 2 times, so the result is not a one-off.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
| tested by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 2/3 | Runs with mocks | 92.00 | 0.06 | |
| Argusic Runner | 2/3 | Runs | 100.00 | 0.08 |
Topics (from GitHub)
accountingai-analysiscurrency-exchangeexpensesgeminiinvoicesllmocr-recognitionollamaopenaiself-hostedtaxes
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/taxhacker)Questions
- Does TaxHacker run?
- Yes. TaxHacker runs. Argusic installed and launched it on a clean machine in 14 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test TaxHacker?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 6cb7254b854b. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test TaxHacker?
- The run that produced this verdict cost $0.08: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does TaxHacker take to install?
- 14 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does TaxHacker need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing TaxHacker?
- 3 things broke in the recorded run, and 3 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for TaxHacker?
- All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.