testing-patterns

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • [Prompt Injection] (SAFE): No instructions designed to override agent behavior or bypass safety filters were found. The content is strictly educational.
  • [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths, or network exfiltration patterns are present.
  • [Obfuscation] (SAFE): No encoded strings, hidden characters, or homoglyphs detected.
  • [Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill does not download or execute external scripts or packages.
  • [Indirect Prompt Injection] (SAFE): While the skill definition includes tools like Bash and Write, the content itself is static documentation and does not ingest or process untrusted external data.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 12:44 PM