qa-testing-cross-browser-compatibility

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides CLI commands and Python scripts intended for execution to generate test matrices and capture screenshots, which is normal behavior for a testing automation tool.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection via external web content captured during the testing process.
  • Ingestion points: The scripts/matrix_runner.py script retrieves the <title> of visited web pages using the page.title() method.
  • Boundary markers: Captured metadata is stored in manifest files without explicit delimiters or safety instructions to distinguish it from agent instructions.
  • Capability inventory: The script has capabilities for browser automation and file system writes (screenshots and JSON manifests) in scripts/matrix_runner.py.
  • Sanitization: The captured page titles are logged and stored without sanitization or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 03:29 AM
Security Audit — agent-trust-hub — qa-testing-cross-browser-compatibility