taste-skill
Agent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
Runssource: GitHubhomepageJavaScriptMITcommit a60a36dc2bb3
JavaScript, MIT licensed. The project labels itself: agent skills, ai agents, anti slop, claude code, codex, cursor, design system and frontend.
taste-skill runs. An Argusic agent installed it in 15 minutes and hit 1 error and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
The taste-skill bundle installs, its 5 tests pass, draw/lint/assets work, and inspect.mjs now runs a real headless Chrome and measures pages, returning a fail/warn list.
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 100 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.
- inspect.mjs aborts with extracted Chrome: SUID chrome-sandbox not configured', so the inspector exited. The CDP client spawns Chrome without --no-sandbox.4 minutes
- Install time
- 15 minutes
- Cold machine to finish
- 4 minutes
- Errors hit and fixed
- 1 hit, 1 fixed with no human help
- How the result was proved
- bash tests/skill_test.sh exited 0 (5/5 ok); install.sh --codex linked 14 skills; draw.mjs wrote direction.json; assets.mjs downloaded a real 1920x1305 image with credits; inspect.mjs in real Chrome returned 'RESULT: 3 FAIL, 3 WARN'; 6 screenshots decoded with ffmpeg -v error
- Model tokens used
- 328,487
- Exact commit tested
- a60a36dc2bb3
- 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.
- 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.
- `./install.sh --list` → 14 install names printed, exit 0. `./install.sh --codex` → 14 symlinks cr… - `node skills/taste-skill/scripts/draw.mjs --sector restaurant --brief "…" --seed 42 --out .taste-… - `node skills/taste-skill/scripts/assets.mjs "Alfama Lisbon old street" --source commons --license… - `node skills/taste-skill/scripts/lint.mjs page-test` → "RESULT: 0 FAIL, 0 WARN", exit 0. - `inspect.mjs` under real headless Chrome 155 (Node 22) → "RESULT: 3 FAIL, 3 WARN — not done." wit… - Business-critical finding: the produced `direction.json` had `"allow": []` and, after a subsequen… **Error found and fixed** - `inspect.mjs` with the downloaded Chrome aborted: Chrome 155's SUID `chrome-sandbox` (extracted w… **Environment gaps recorded** - Root installs are impossible (no root, `apt-get install` shows `E: Could not open lock file`), so… - Node 18 is the system default; `inspect.mjs` needs Node 22+ (`typeof WebSocket` check). I used `/… - The only supporting files I left in the working tree are the harness's own two eval dirs, `page-t…
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.
- Recovered from all 1 error 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.
What did not, or is not known
- Took 15 minutes to install, slower than the median of the 59 comparable projects Argusic has measured.
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Of the 60 agent-skills projects Argusic has installed and timed, taste-skill was the 53rd fastest to reach a running state, and 45 of 60 reached one at all.
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
Topics (from GitHub)
agent-skillsai-agentsanti-slopclaude-codecodexcursordesign-systemfrontendlanding-pageui-designweb-design
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/taste-skill)Questions
- Does taste-skill run?
- Yes. taste-skill runs. Argusic installed and launched it on a clean machine in 15 minutes, hitting 1 error on the way, and recorded the session.
- How did Argusic test taste-skill?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit a60a36dc2bb3. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test taste-skill?
- The run that produced this verdict cost $0.06: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does taste-skill take to install?
- 15 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 taste-skill need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing taste-skill?
- 1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does taste-skill compare with the alternatives?
- Of the 60 agent-skills projects Argusic has installed and timed, taste-skill was the 53rd fastest to reach a running state, and 45 of 60 reached one at all.
- Where is the evidence for taste-skill?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.