playwright-scraper-skill

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified. The skill's operations, including network access and file system writes, are consistent with its purpose as a web scraping tool.
  • [COMMAND_EXECUTION]: The skill utilizes Playwright to launch a Chromium browser and execute extraction scripts, which is standard behavior for browser automation tools.
  • [PROMPT_INJECTION]: The skill represents an ingestion surface for indirect prompt injection as it processes untrusted data from external websites.
  • Ingestion points: Web content is fetched and extracted in scripts/playwright-simple.js (line 39) and scripts/playwright-stealth.js (line 122) via page.evaluate().
  • Boundary markers: Scraped content is returned within a JSON structure without specific delimiters or isolation markers.
  • Capability inventory: The skill can perform network requests (page.goto) and write files to the filesystem (screenshots and HTML saves).
  • Sanitization: Extracted text and HTML content are not sanitized or filtered before being returned.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 05:10 PM