The Agent Skills Directory

[SAFE]: The skill includes a protective blacklist to prevent the accidental staging of sensitive files such as credentials, environment variables, and private keys.
[SAFE]: Execution flow includes human-in-the-loop validation, requiring user approval via the 'AskUserQuestion' tool before performing 'git add' or 'git commit' operations.
[SAFE]: The skill uses a quoted HEREDOC for git commit messages, which is a secure method to prevent command injection from generated content.
[SAFE]: Tool access is restricted to local git operations, with no network access or external dependency fetching identified.

create-commit