web-visual-verification

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill generates and facilitates the execution of JavaScript-based automation scripts using the Node.js runtime environment.\n- [EXTERNAL_DOWNLOADS]: Functional templates rely on the well-known playwright library for cross-browser testing and automation.\n- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes untrusted content from external websites.\n
  • Ingestion points: External data enters the agent context via page.goto() and page.evaluate() operations within the generated Playwright scripts in SKILL.md.\n
  • Boundary markers: No specific boundary markers or instructions to ignore embedded content are used when scraping or interacting with site data.\n
  • Capability inventory: The skill has the ability to write to the local file system using node:fs (for directory creation) and page.screenshot() (for image capture), alongside standard browser control capabilities.\n
  • Sanitization: No evidence of data sanitization or validation of the content retrieved from external URLs is present.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 01:17 PM
Security Audit — agent-trust-hub — web-visual-verification