README.md

metadata

license: apache-2.0
pretty_name: gLabels computer-use environment from a one-line request (REA)
tags:
  - computer-use
  - cua
  - gui-agent
  - rl-environment
  - agentenv
  - desktop
  - linux
  - rea
  - labels
  - label-design
  - printing
  - from-request

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

The real gLabels 3.4 label designer 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. gLabels opens a one-label document, Documents/address-label.glabels, at launch: an Avery 5160 address label for Marisol Penhalow, 418 Juniper Court, Bellmont, OR 97215. The agent has to correct the name to Marisol Penhallow and save. 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

The name on my address label in Documents/address-label.glabels is misspelled. It should read Marisol Penhallow. Fix the name in gLabels and save the label.

bundle/artifacts/fix-label-name-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
check-1	text_equals	Documents/address-label.glabels

A .glabels file is gzip-compressed XML, and REA reads the XML inside. The check reads the whole text of the label's text object, the Span inside Object-text, where gLabels writes each line break as an empty NL element, so the lines run together: it must equal "Marisol Penhallow418 Juniper CourtBellmont, OR 97215" exactly, which proves the name is fixed and the address untouched. It is the one check REA's draft suggested, kept exactly as written.

Results

3 of 3 runs passed, about $0.68 per run.

Run	Outcome	Turns	Cost	Video
1	passed	10	$0.98	video
2	passed	10	$0.46	video
3	passed	17	$0.59	video

Run it

You need Docker. agentenv-framework builds the computer: Ubuntu 24.04 with the app installed from Ubuntu's archive (glabels 3.4.1, with librsvg2-common).

pip install agentenv-framework==0.9.1431
hf download earakely-scale/gLabelsFromRequestComputerEnv-REA --repo-type dataset --local-dir glabels-request
agent-env run glabels-request/bundle --task fix-label-name --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 gLabels environment where the agent fixes the name on a small address label and saves it.

It asked only the skill's upfront questions, wrote no app-specific scripts, and finished in about 30 minutes.

gLabels 3.4.1 crashed the first time its Text tool was clicked in a fresh profile; Claude Code set a default font and captured again, and the seed keeps that setting among its dconf keys. The environment was named glabels at export, because the build computer's own default name, taken from the desktop entry glabels-3.0, came out as "0". The steps were REA's usual ones:

1. Capture. REA launched gLabels 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 edit was made through computer actions alone, and the checks were read at four stages. They held on the first try:

Stage	Checks
seed	fail, as expected: the name reads Marisol Penhalow
after_edit	pass
near_miss	fail, as expected: the name was cut to "Marisol Penhallo"
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

• The check reads the label's one text object, name and address together, and needs it exact: any change to the address lines fails, and so does a fix made by adding a second text object.
• gLabels keeps its settings in dconf keys, which the computer loads once when the app first starts; data/reset restores the label but not those settings.
• The bundle installs librsvg2-common, the SVG image loader gLabels' icons need, beside the app as an extra package.
• The computer installs whatever glabels Ubuntu 24.04's archive holds at build time (3.4.1 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. gLabels itself is not included: the computer installs it from Ubuntu's archive (glabels 3.4.1, with librsvg2-common) at build time, under its own licence (GPL-3.0-or-later).