Baguette is a command-line tool and web interface for controlling iOS simulators without opening Xcode or Simulator.app. Developers and agents use it to boot devices, stream screens, send touch and hardware input, inspect accessibility data and logs, and capture screenshots or recordings. The catalogue includes skills, instructions, and a plugin for operating Baguette.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add tddworks/baguette --skill baguette-test-appgit clone --depth 1 https://github.com/tddworks/baguetteWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/tddworks/baguette/baguette-test-app)<a href="https://agentmods.dev/skills/tddworks/baguette/baguette-test-app"><img src="https://agentmods.dev/badge/skills/tddworks/baguette/baguette-test-app/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tddworks/baguette/baguette-test-app"><img src="https://agentmods.dev/badge/skills/tddworks/baguette/baguette-test-app.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00118 | $0.01937 |
| Opus 5.5 | $0.00047 | $0.00775 |
| Sonnet 5.5 | $0.00024 | $0.00387 |
| Haiku 4.5 | $0.00012 | $0.00194 |
Grade A, and why
baguette-test-app scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Test an app with baguette
The test is a loop you do yourself, not a script you write:
read the screen → think what's next → click → check it worked.
Your judgment is the point. A script can only find what it was told to look for; you can notice that the price is wrong.
Before you start
baguette list --json # a booted UDID
baguette boot --udid <X> # if none is running
xcrun simctl install <X> path/to/App.app # if it isn't installed
If the user wants to watch, start baguette serve and give them
http://localhost:8421/simulators/<X>.
Write down what success looks like before the first click, in the user's words. "Signed up" isn't enough: "a new account, on the free plan, with the receipt emailed".
The loop
1. Open the app and read the screen, in one call.
baguette launch --udid <X> --bundle-id <id> --relaunch --then-describe-ui
It waits for the first real screen, usually 4 to 5 seconds. Baguette logs to stderr,
so add 2>/dev/null if you only want the answer. This first answer also has how, the
commands to use from here. Answers to a tap leave it out, because you just ran one.
2. Read the answer and think. It's short, and says what you can do:
{"ok":true,"changed":true,
"affordances":{"tap":["Back","Groceries, 3 items","Holiday, 12 items","New List"]},
"disabled":["Share List"],
"text":["My Lists"],
"values":{"Home":"1"}}
affordances.tap: the controls you can tap, by name.affordances.type: a focused text field you can type into.affordances.adjust: a slider. A press can't set a value, so drag it withswipe, using its frame fromdescribe-ui.disabled: controls that can't be used yet. Do what they're waiting for first.text: what the screen says. Some of it can be tapped too, like aShow morelink.values: a control's value, like a switch's1or the selected tab's.
Read what's there before choosing: the first row is rarely the one the task means.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 176 lines · 118 tokens per session scan A c2b33c596a75
baguette-test-app is a skill published in the GitHub repository tddworks/baguette (2,273 stars, last pushed yesterday), licensed Apache-2.0. It adds 118 tokens to every session and 1,937 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-06.
baguette-test-app compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| baguette-test-apptddworks/baguette | Skill | 2.3k | yesterday | Apache-2.0 | A | 118 | Claude Code |
| mobile-automationmobile-next/mobile-mcp | Skill | 10k | 4d ago | Apache-2.0 | A | 58 | Claude Code, Codex |
| Detox Mobile TestingPramodDutta/qaskills | Skill | 245 | 3d ago | MIT | A | 48 | Claude Code, Codex |
| noqa-testingnoqa-ai/noqa | Skill | 191 | 3mo ago | MIT | A | 47 | Claude Code, Codex |
| xcforgejustinthevoid/xcforge | Skill | 2 | 27d ago | MIT | A | 140 | Claude Code, Codex |
| cmp-qa-prepkvdm-co-pilot/create-cmp | Skill | 0 | 16d ago | MIT | C | 131 | Claude Code |
| argent-test-ui-flowsoftware-mansion/argent | Skill | 3.2k | yesterday | Apache-2.0 | A | 61 | Claude Code, Codex |