qa-testing

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a procedural guide for software testing, focusing on data integrity and application stability. It utilizes standard developer tools such as npm and Playwright for legitimate testing purposes.
  • [COMMAND_EXECUTION]: The skill includes shell commands for running test suites (e.g., "npm test", "npm run test:stress:full"). These are standard development operations intended for local testing and validation within a project environment.
  • [DATA_EXPOSURE]: While the skill interacts with a database (Supabase), it does so using instructions and code snippets for verification purposes during testing. No hardcoded credentials or unauthorized data exfiltration patterns were found.
  • [PROMPT_INJECTION]: The skill contains instructional markers such as "CRITICAL" or "ALWAYS", but these are used in the context of coding best practices (e.g., cleaning up event listeners) rather than attempts to bypass agent safety filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 01:14 AM
Security Audit — agent-trust-hub — qa-testing