commit-and-pr
Pass
Audited by Gen Agent Trust Hub on Mar 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for routine developer workflows and incorporates safety rules to prevent common mistakes, such as force-pushing to protected branches.
- [COMMAND_EXECUTION]: The skill executes Git and GitHub CLI commands. It uses secure shell patterns like heredocs (
<<'EOF') for commit messages and pull request bodies, which prevents the content from being incorrectly interpreted as shell commands. - [DATA_EXFILTRATION]: While the skill pushes code to remote repositories, it contains explicit instructions to avoid staging sensitive files like
.env, secrets, or authentication tokens. - [PROMPT_INJECTION]: The skill processes file differences (
git diff) to generate commit messages and PR descriptions. This is an indirect injection surface, but the skill's structure isolates the generated text from the command execution environment using temporary files and heredocs, mitigating the risk.
Audit Metadata