ui-demo

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's instructions and code snippets align with its stated purpose of browser automation and video recording. No obfuscation, persistence mechanisms, or credential theft patterns were found.
  • [COMMAND_EXECUTION]: The skill facilitates the creation and execution of Node.js scripts using Playwright. This involves launching browser instances (e.g., Chromium) and performing automated interactions based on a generated script.
  • [EXTERNAL_DOWNLOADS]: The skill references the 'playwright' Node.js package, which is an industry-standard library for web automation.
  • [DATA_EXFILTRATION]: The skill ingests metadata from web pages to assist the agent in identifying interactive elements.
  • Ingestion points: Page elements (tags, types, names, placeholders, and text content) are scraped using page.evaluate in the 'Phase 1: Discover' section.
  • Boundary markers: None observed; the skill relies on direct DOM inspection.
  • Capability inventory: The skill uses Playwright for browser control and the Node.js fs module for local video file management.
  • Sanitization: Scraped element data is logged directly to the console for the agent to use when writing scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 09:46 AM