compatibility-testing

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Data Exposure & Exfiltration] (SAFE): The skill demonstrates secure credential management by using environment variables (process.env.BROWSERSTACK_USER) instead of hardcoding API keys or secrets.
  • [Unverifiable Dependencies] (SAFE): The skill utilizes @playwright/test, a trusted library from a known organization (Microsoft), for its automation logic.
  • [Indirect Prompt Injection] (SAFE): There is an inherent surface for indirect injection when the agent navigates to external URLs for testing. However, as this is the primary intended purpose of a compatibility tester, the risk is considered low.
  • Ingestion points: External URLs processed via page.goto and the Task definition in SKILL.md.
  • Boundary markers: Absent.
  • Capability inventory: Browser navigation and visual screenshot comparison via qe-visual-tester.
  • Sanitization: None specified in the snippets, relying on Playwright's default sandbox behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:48 PM