caveman-commit
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed entirely of natural language instructions and formatting rules for commit messages. It does not include any scripts, executable code, or remote dependencies.- [COMMAND_EXECUTION]: The instructions include a 'Boundaries' section that strictly limits the agent to text generation and explicitly forbids running
git commit,git stage, or other shell commands.- [PROMPT_INJECTION]: The skill was analyzed for indirect prompt injection risks associated with processing git diffs. Given the lack of tool access or execution capabilities, this surface presents no significant security risk.
Audit Metadata