codebase-exploration

Pass

Audited by Gen Agent Trust Hub on Jul 4, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is defined as an 'agent-guide' and consists entirely of markdown instructions. It contains no executable code or scripts.
  • [PROMPT_INJECTION]: No malicious instruction overrides, role-play bypasses, or safety guideline violations were detected in the instructions.
  • [DATA_EXFILTRATION]: The skill does not include any commands for network communication, data exfiltration, or access to sensitive local credential files.
  • [REMOTE_CODE_EXECUTION]: No patterns were found for downloading or executing remote code, scripts, or unverifiable packages.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 4, 2026, 10:55 PM
Security Audit — agent-trust-hub — codebase-exploration