macos-sandbox

macos-sandbox is a skill for Claude Code from robzilla1738/roberts-skills. It costs 61 tokens per session (1,134 once invoked), scanned A, original, MIT.

A workflow for testing a macOS app or installer inside a disposable local virtual machine, which is an isolated software computer.

In plain words
What is it for?
Use it with macbox to launch .app or .pkg builds, capture screenshots, inspect logs and crash reports, automate the guest, and reset the test machine.
Why use it?
It provides a clean environment for finding launch failures, collecting evidence, and testing without affecting the host Mac.

Skill for Claude Code

Written for Claude Code: disable-model-invocation in frontmatter.

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /Users/admin/Desktop/MyApp.app.

Part of the macos-sandbox plugin — 1 skill, 1 command shipped together

Good fit Use it with macbox to launch .app or .pkg builds, capture screenshots, inspect logs and crash reports, automate the guest, and reset the test machine.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add robzilla1738/roberts-skills
Claude Code
/plugin install macos-sandbox

Made for: Claude Code.

Or install macos-sandbox, the plugin that ships this one along with the rest of its 1 skill, 1 command.

Wrote 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.

agentmods badge for macos-sandbox

README.md
[![agentmods](https://agentmods.dev/badge/skills/robzilla1738/roberts-skills/macos-sandbox/github.svg)](https://agentmods.dev/skills/robzilla1738/roberts-skills/macos-sandbox)
Your own site
<a href="https://agentmods.dev/skills/robzilla1738/roberts-skills/macos-sandbox"><img src="https://agentmods.dev/badge/skills/robzilla1738/roberts-skills/macos-sandbox/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.

agentmods 80×15 button for macos-sandbox

Your own site · 80×15
<a href="https://agentmods.dev/skills/robzilla1738/roberts-skills/macos-sandbox"><img src="https://agentmods.dev/badge/skills/robzilla1738/roberts-skills/macos-sandbox.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 61 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,134 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00061 $0.01134
Opus 5 $0.00030 $0.00567
Sonnet 5 $0.00012 $0.00227
Haiku 4.5 $0.00006 $0.00113

Measured 10d ago against content hash 814843e57b32, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

macos-sandbox 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 10d 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.

plugins/macos-sandbox/skills/macos-sandbox/SKILL.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

macOS Sandbox Testing with macbox

Use macbox when you need to validate a macOS app build in a clean local VM without touching the host.

When to use

  • Smoke-test a freshly built .app or .pkg
  • Capture launch screenshots, system logs, and crash reports
  • Reproduce startup failures in an isolated macOS guest
  • Reset or destroy a sandbox after testing

When not to use

  • Do not use macbox for cloud deployment or CI farm orchestration
  • Do not upload secrets, credentials, SSH keys, or browser profiles
  • Do not expect macbox to bypass Gatekeeper, Keychain, or other security prompts

Prerequisites

  1. Host has Tart installed and a prepared base image (for example macos-sequoia-clean)
  2. SSH key auth works: ~/.ssh/macbox_id installed in guest admin account
  3. MCP server configured locally (stdio only) or CLI available as macbox

Safe workflow

1. Check readiness

macbox doctor --json
macbox status --json

Confirm doctor reports Tart, ssh/scp, SSH identity, and the state directory as OK.

2. Create sandbox

Via MCP: create_sandbox(image="macos-sequoia-clean", headless=True)

Via CLI:

macbox start --image macos-sequoia-clean --name macbox-test-001 --headless --json

Save the returned vm name and run_id.

3. Upload build artifact

Upload only .app or .pkg files from explicit local paths.

Via MCP: upload_app(vm_name="macbox-test-001", app_path="/path/to/MyApp.app")

Via CLI:

macbox upload --name macbox-test-001 --path ./dist/MyApp.app --dest /Users/admin/Desktop/MyApp.app --json

4. Run smoke test

Via MCP: run_app_smoke_test(vm_name="macbox-test-001", app_name="MyApp.app", timeout_seconds=120)

Via CLI:

macbox run-app --name macbox-test-001 --app /Users/admin/Desktop/MyApp.app --timeout 120 --json

Inspect JSON data:

  • launched
  • crashed
  • screenshot
  • logs
  • crash_reports

5. Collect additional evidence if needed

macbox logs --name macbox-test-001 --last 5m --json
macbox screenshot --name macbox-test-001 --json
macbox collect-crashes --name macbox-test-001 --json

Read the full file on GitHub · 159 lines

Files

What ships with it

1 file 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.

Changes

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.

  1. 10d ago First seen · 159 lines · 61 tokens per session scan A 814843e57b32

Subscribe to this mod's changes

macos-sandbox is a skill published in the GitHub repository robzilla1738/roberts-skills (5 stars, last pushed 3mo ago), licensed MIT. It adds 61 tokens to every session and 1,134 once invoked, about $0.0003 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 2026-08-31.

Related

Other skills, from other repositories

hatch3r-browser-verify

Opt-in browser verification skill — spec-run-first Playwright verification (assertions execute in the runner, agent reads only failures), axe-core a11y audits, toHaveScreenshot() regression diffs, E2E test scaffolds, and snapshot-mode exploratory driving. Default ON for UI-affecting agent invocations; disable globally…

hatch3r/hatch3r · 83 tokens

hatch3r-qa-path

Produces a human-run manual QA test path from a PR, branch diff, or uncommitted working tree — a risk-ordered table of steps, expected results, and automated-coverage references plus a shippability sign-off. Rows are auto-proven via delegated verification sub-agents before emission, so the human walks only what…

hatch3r/hatch3r · 100 tokens

hatch3r-a11y-audit

Runs a WCAG AA accessibility audit with findings and fixes across 7 scan categories (keyboard, contrast, ARIA, reduced motion, screen reader, high contrast, automated axe). Use when auditing and remediating accessibility.

hatch3r/hatch3r · 55 tokens

hatch3r-qa-validation

E2E validation workflow producing a structured pass/fail report with evidence. Use when running QA validation, acceptance testing, verifying releases, or working on QA E2E validation issues.

hatch3r/hatch3r · 45 tokens

browser-qa

Use this skill to automate visual testing and UI interaction verification using browser automation after deploying features.

sifxprime/kodelyth-ecc · 22 tokens

xcode-ui-test-stabilizer

Build and stabilize Xcode UI end-to-end tests with XCUIApplication/xcodebuild for new or unreliable automation, covering environment setup, focus/input reliability, waits, logs, attachments, and flakiness triage.

Xopoko/build-swift-apps · 51 tokens