earakely-scale/GnuCashFromRequestComputerEnv-REA

GnuCash: a computer-use environment built from a one-line request

The real GnuCash 5.5 accounting app running on an Ubuntu 24.04 desktop, as an agentenv computer environment. An agent completes the task only through screenshots, mouse and keyboard, and a verifier grades the saved files with no model.

Claude Code built this environment from a one-line request in about 30 minutes, with only REA's build skill and MCP tools; How it was made has the request and the steps. GnuCash opens a small synthetic household book, Documents/household.gnucash, at launch, behind its Tip of the Day dialog: a standard set of 64 accounts and four October transactions in the checking account, among them a Corner Grocer purchase. The agent has to record a $63.48 purchase at Greenleaf Market, paid from Checking Account on November 3, 2026, under Expenses:Groceries, and save the book. It sees only the screen and acts only with the mouse and keyboard.

Watch

• Agent runs (Claude Code with Claude Sonnet 5.5): run 1, run 2, run 3
• Witness, which proves the task can be done through the computer: witness.mp4

Each video is the computer's own screen recording at 10 frames per second.

The task

In my household book (Documents/household.gnucash), record a grocery purchase: on November 3, 2026 I paid $63.48 from the Checking Account at Greenleaf Market. Enter it with the description Greenleaf Market and file it under Expenses:Groceries, then save the book.

bundle/artifacts/record-grocery-purchase-check/verify.py reads the app's files through the computer's data/get view and passes only if every check holds:

Check	Kind	File
greenleaf-recorded-once	member_count	Documents/household.gnucash
groceries-debited	text_equals	Documents/household.gnucash
checking-credited	text_equals	Documents/household.gnucash
posted-nov-3	text_contains	Documents/household.gnucash
one-transaction-added	member_count	Documents/household.gnucash
corner-grocer-kept	text_equals	Documents/household.gnucash
guard-1	text_equals	Documents/household.gnucash

A .gnucash file is GnuCash's XML, gzip-compressed by default, and REA reads the XML inside. The checks find the new transaction by its description, Greenleaf Market, with a selector on a child element's text ([>trn:description=Greenleaf Market]), never by position: exactly one transaction must have it, its Expenses:Groceries split must be 6348/100 and its Checking Account split -6348/100 (GnuCash writes amounts as fractions), and it must be posted on 2026-11-03. The book must hold exactly five transactions, the Corner Grocer purchase must keep its 52.75, and the book must keep its 64 accounts. REA's draft suggested nine checks, mostly on the book's bookkeeping slots, ids and entry times; Claude Code edited four, discarded five, and took the two accounts' GUIDs from REA's own export refusals.

Results

3 of 3 runs passed, about $0.63 per run.

Run	Outcome	Turns	Cost	Video
1	passed	11	$1.03	video
2	passed	12	$0.32	video
3	passed	11	$0.55	video

Run it

You need Docker. agentenv-framework builds the computer: Ubuntu 24.04 with the app installed from Ubuntu's archive (gnucash 5.5).

pip install agentenv-framework==0.9.1431
hf download earakely-scale/GnuCashFromRequestComputerEnv-REA --repo-type dataset --local-dir gnucash-request
agent-env run gnucash-request/bundle --task record-grocery-purchase --agent claude-code --model <model>

The agent needs a model provider configured for agentenv. --task smoke only boots the computer, with no agent. bundle/README.md describes the bundle in detail.

How it was made

Claude Code built this environment from a one-line request, given only REA's build skill and REA's MCP tools:

Build me a GnuCash environment where the agent records a grocery purchase in a small household book and saves it.

It asked only the skill's upfront questions, wrote no app-specific scripts, and finished in about 30 minutes. The steps were REA's usual ones:

1. Capture. REA launched GnuCash on a Linux desktop and recorded its package, windows and opened files while the edit was made. It also recorded the app's state folders before and after.
2. Export. REA turned the capture into an agentenv computer_app. It installs the app, seeds the starting files, starts the app, and waits for its main window. It also serves typed views of the app's files for verifiers.
3. Task. The task's checks were validated against the capture: they fail on the starting state and pass on the captured end state.
4. Witness. In a fresh computer, the purchase was recorded through computer actions alone, and the checks were read at four stages. They held on the first try:

Stage	Checks
seed	fail, as expected: there is no Greenleaf Market transaction; the kept checks pass
after_edit	pass
near_miss	fail, as expected: the amount was changed to 63.84
after_reset	fail, as expected: data/reset restored the seed

The agent runs came after the build: Claude Code with Claude Sonnet 5.5 (--model anthropic/claude-sonnet-5-5 --no-emulate, on an amd64 host) solved the task 3 times, each run recorded.

What's inside

bundle/      the agentenv bundle: envs/ (app module, seed files), tasks/, artifacts/ (verifier), evals/
media/       witness and agent-run recordings (MP4) with short GIF previews
results.json the runs above, with each check's result

Limits

• GnuCash shows its Tip of the Day dialog at every boot and after every data/reset, and the agent has to close it: GnuCash keeps that setting in dconf, which this build did not seed.
• The split checks name the Groceries and Checking Account accounts by GUID. These are the seed's own ids, so they hold in every computer, but a purchase filed under an account the agent creates, even one named Groceries, fails.
• The checks find the transaction by its description, so a purchase entered under another description, such as "Greenleaf", is not found and every transaction check fails.
• The seed also carries a stray transaction log, ~/translog.20261106084748.log, which GnuCash wrote in the home folder while the book was being made; the bundle owns it as a path. The checks read only the book.
• The computer installs whatever gnucash Ubuntu 24.04's archive holds at build time (5.5 for these runs); it is not pinned.
• Three runs is a small sample: the solve rate shows the task is solvable, not how hard it is.

Licences

The bundle code, the task and the synthetic seed files are Apache-2.0. GnuCash itself is not included: the computer installs it from Ubuntu's archive (gnucash 5.5) at build time, under its own licence (GPL-2.0-or-later).