playwright-recording

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill automates browser navigation to external websites which could contain malicious content or instructions. * Ingestion points: page.goto(url) and page.goto(CONFIG.url) calls in SKILL.md and reference.md. * Boundary markers: No delimiters or warnings are provided to the agent regarding instructions potentially embedded in the targeted web pages. * Capability inventory: The skill utilizes fs for file manipulation, execSync for shell command execution, and Playwright for full browser control. * Sanitization: No sanitization or filtering of external web content is performed.
  • [COMMAND_EXECUTION]: The skill uses shell commands to process video files. * Uses child_process.execSync in reference.md to execute ffprobe for duration calculation. * Suggests the use of ffmpeg in SKILL.md for video conversion.
  • [EXTERNAL_DOWNLOADS]: The skill fetches external packages and browser binaries from well-known official sources. * npm install -D playwright @playwright/test installs standard browser automation libraries. * npx playwright install chromium downloads Chromium binaries from the Playwright official repository.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 06:31 PM