git-commit-conventions
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines best practices for git workflows and does not contain any evidence of prompt injection, data exfiltration, or unauthorized code execution.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local repository and executes git commands based on that content, creating a potential attack surface.
- Ingestion points: Local repository files, git diff output, and recent commit history (SKILL.md).
- Boundary markers: No explicit boundary markers or instructions to ignore embedded content are present.
- Capability inventory: Access to git commands (add, commit, reset, show, stash, diff) and file system operations (SKILL.md).
- Sanitization: No content sanitization or validation of repository data is specified.
Audit Metadata