create-apple-notes-mockup

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses Playwright to manage a headless Chromium instance for screenshot generation. The package.json specifies playwright as a dependency, and the SKILL.md setup instructions include npx playwright install chromium. These are standard requirements for this library and originate from a well-known service.
  • [COMMAND_EXECUTION]: The skill includes a tests/run-all.sh script and a render.js CLI. These tools are used for the primary purpose of the skill: rendering mockups and running visual regression tests. The shell commands are static and do not interpolate untrusted data in a dangerous way.
  • [DATA_EXPOSURE]: The skill processes local images provided in the JSON spec by reading them from disk and inlining them as Base64 data URIs into the generated HTML. This is restricted to the local file system relative to the spec file and is a core functional requirement.
  • [PROMPT_INJECTION]: As a deterministic renderer that does not involve LLM processing of its own instructions, there is no surface for direct prompt injection. The skill documentation correctly identifies that the 'orchestrating agent owns the copy' while this skill owns the fidelity.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes JSON data which is then interpolated into HTML. The implementation in generate.js uses a robust escapeHtml function and smartQuotes processing to sanitize text before it is inserted into the HTML template, preventing XSS or structural confusion in the rendered output.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 04:39 PM
Security Audit — agent-trust-hub — create-apple-notes-mockup