git-commit
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes repository diffs and status to generate commit messages. This is an inherent attack surface for indirect prompt injection if the code being committed contains adversarial instructions.
- Ingestion points: Local file diffs and git status (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: Executes standard
gitcommands likeaddandcommit(SKILL.md). - Sanitization: Absent.
- [SAFE]: The skill instructions focus on standard Git workflows and explicitly discourage dangerous operations like force pushing or bypassing pre-commit hooks, ensuring that Git hooks are allowed to execute normally.
Audit Metadata