The Agent Skills Directory

[SAFE]: The skill implements a developer-centric workflow using standard git commands (status, diff, log, add, commit) for repository management. No unauthorized or suspicious command execution patterns were found.
[SAFE]: The included helper script scripts/git-info.sh gathers repository metadata and formats it as JSON using jq. It does not perform any network operations, access sensitive user files (like SSH keys), or use dangerous functions like eval or exec on external input.
[SAFE]: Explicit guardrails are included in the instructions to protect the user environment, such as directing the agent to stop if potential secrets are found in the diff and prohibiting dangerous flags like --no-verify or force-pushing.
[SAFE]: No obfuscation, prompt injection, or remote code execution patterns were detected in the skill instructions or associated scripts.

commit