The Agent Skills Directory

[SAFE]: The skill explicitly enforces security best practices, such as using environment variables for credentials and prohibiting the commitment of secrets or API keys to repository history.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform standard Git and GitHub CLI operations. Execution is bounded by strict behavioral constraints (e.g., no direct pushes to protected branches) and requires the agent to articulate intent before performing high-risk actions.
[PROMPT_INJECTION]: The skill handles data from external repositories, creating a surface for indirect prompt injection. However, it mitigates this risk by requiring manual verification of diffs and providing a security checklist for the agent to follow before finalizing operations.

github-operations