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 MadAppGang/magus --skill macos-app-testing-no-disruptgit clone --depth 1 https://github.com/MadAppGang/magusWrote 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/madappgang/magus/macos-app-testing-no-disrupt)<a href="https://agentmods.dev/skills/madappgang/magus/macos-app-testing-no-disrupt"><img src="https://agentmods.dev/badge/skills/madappgang/magus/macos-app-testing-no-disrupt/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/madappgang/magus/macos-app-testing-no-disrupt"><img src="https://agentmods.dev/badge/skills/madappgang/magus/macos-app-testing-no-disrupt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00200 | $0.03997 |
| Opus 5 | $0.00100 | $0.01998 |
| Sonnet 5 | $0.00040 | $0.00799 |
| Haiku 4.5 | $0.00020 | $0.00400 |
Grade A, and why
macos-app-testing-no-disrupt 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 7d ago.
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 — 304 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Testing native macOS apps without disrupting the user
The core problem
You need to confirm a native macOS app does what it should — see its window,
check what it wrote to disk, verify a process spawned — but the user is sitting
at the same Mac, working. The naive approach (computer-use screenshot + clicks)
forces the target app to the foreground, hijacking the user's mouse, keyboard,
and focus. That is the disruption to avoid.
The insight that makes this tractable: macOS has two separate input paths, and only one of them needs the app frontmost.
- You can see any window's pixels without focusing it —
screencapture -lreads a single window's backing store directly, even when it's buried behind other windows. - You can click and type into a backgrounded app via the Accessibility
API (
AXPress,set value). AX delivers a semantic action straight to the app's element tree, bypassing the cursor and the frontmost-app event routing — so focus never moves. What you cannot do focus-free is synthesize a hardware mouse/keyboard event at a screen coordinate (computer-use, rawcliclick) — those go to whatever's frontmost, so they require foregrounding.
So the disruption-free strategy is: observe by capture+inspect, and DRIVE by AX — both without taking focus. Reach for the human only when a specific control genuinely won't expose to AX (see "Clicking … the AX way"). And lead with side-effect checks where you can: much verification needs no clicks at all, because a native app's behavior is written into files, processes, and logs you can read without touching its window.
Decision: do you even need the GUI?
Before reaching for a screenshot, ask what you're actually verifying. A native app's real behavior is usually observable without its window:
- Did it write the right config / output? → read the files it writes (capture a baseline first; see "Verify by side effect").
- Did it spawn the right helper / server? →
pgrep -fl, check the port. - Did it log success / error? → tail its log file.
- Does the UI actually render X? (a label, a picker entry, a state) → this is what window capture is for.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 7d ago First seen · 304 lines · 200 tokens per session scan A b92f0d311887
macos-app-testing-no-disrupt is a skill published in the GitHub repository MadAppGang/magus (9 stars, last pushed today), licensed MIT. It adds 200 tokens to every session and 3,997 once invoked, about $0.0010 per session on Opus 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 2026-09-04.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.