explore-code

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured guidance for understanding unfamiliar codebases using standard documentation and code analysis techniques.
  • [COMMAND_EXECUTION]: The skill uses the 'tree' command to provide a visual map of the directory structure. This is a standard, low-risk operation for orientation.
  • [PROMPT_INJECTION]: While the skill's purpose is to read and interpret external (potentially untrusted) code, the instructions themselves do not contain any bypass or override patterns. Users should remain aware that the code being analyzed could contain indirect prompt injections.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 02:50 AM
Security Audit — agent-trust-hub — explore-code