manage-submodules

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides standard procedural guidelines for Git submodule management without introducing dangerous command execution or network operations.
  • [PROMPT_INJECTION]: No behavioral overrides, safety filter bypasses, or instructions to ignore previous rules were detected.
  • [DATA_EXFILTRATION]: No hardcoded credentials or data exfiltration patterns were found. The skill explicitly lists 'secret disclosure' as a prohibited effect.
  • [COMMAND_EXECUTION]: The skill focuses on standard Git workflows and does not provide instructions for executing arbitrary scripts or unauthorized system commands.
  • [INDIRECT_PROMPT_INJECTION]: While the skill involves processing external repository data (such as .gitmodules and commit messages), it includes explicit instructions to treat such external content as untrusted data until verified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 03:46 AM
Security Audit — agent-trust-hub — manage-submodules