conventional-commits

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • [No Code] (SAFE): This skill consists entirely of Markdown instructions and examples. There are no scripts, binaries, or configuration files that execute commands on the host system.
  • [Prompt Injection] (SAFE): The content provides clear formatting rules without attempting to override the agent's core safety guidelines or ignore previous instructions.
  • [Data Exposure & Exfiltration] (SAFE): No network operations, sensitive file path access, or hardcoded credentials were detected.
  • [Indirect Prompt Injection] (INFO): The skill is designed to process external data (git diffs and user-provided descriptions). However, because the skill has no side-effect capabilities (no file writing, no network access, no command execution), the risk of malicious data influencing the system through this skill is negligible.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 09:12 AM