principles

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill incorporates clear safety principles that explicitly forbid the hardcoding of credentials and the modification of protected repository paths (e.g., .github/).\n- [COMMAND_EXECUTION]: The skill guides the agent to use standard local development utilities such as git, npm, npx, ls, and find for repository context gathering and build verification. These operations are appropriate for the skill's primary purpose and do not involve remote execution or unknown sources.\n- [SAFE]: The instructions include specific guidance to avoid reading sensitive environment files or secrets directories, reducing the risk of accidental data exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 11:34 AM