The Agent Skills Directory

[SAFE]: The skill consists of documentation and workflow guidelines. It encourages security best practices such as secret detection before committing and using safer git commands like --force-with-lease.\n- [PROMPT_INJECTION]: No evidence of direct prompt injection or attempts to bypass agent safety filters was found. The skill interacts with user-generated metadata (commit messages, branch names) as part of standard Git operations, but does so through established CLI tools.\n- [DATA_EXFILTRATION]: No hardcoded credentials or unauthorized network operations were detected. The skill proactively includes logic to help users prevent accidental credential exposure during the commit process.\n- [REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts from untrusted sources. It relies on standard, well-known tools like the GitHub CLI (gh) and the git command line interface.

git-workflow