testing-fallback-strategies

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains legitimate instructional content designed to improve the reliability of an AI agent's verification process. No malicious patterns, such as prompt injection, data exfiltration, or obfuscation, were detected.
  • [COMMAND_EXECUTION]: The skill references the use of standard development and testing commands including npm test, npx vitest, npm run build, and npx tsc. These are intended for local environment verification and do not involve untrusted remote sources.
  • [EXTERNAL_DOWNLOADS]: The skill mentions the use of agent-browser, which is a standard tool for browser automation in agentic workflows. It does not initiate unauthorized external downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 02:41 AM
Security Audit — agent-trust-hub — testing-fallback-strategies