conventional-commit

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely informational and serves as a guideline for generating commit messages according to the Conventional Commits specification.
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to analyze the output of git diff --cached to determine the commit type and scope. While the output of a diff could theoretically contain malicious instructions, the skill only uses this data to generate a text-based summary (the commit message) and does not execute the contents of the diff or pass it to dangerous functions. The severity is low as this is a standard workflow for this task.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 11:05 AM
Security Audit — agent-trust-hub — conventional-commit