The Agent Skills Directory

[SAFE]: The instructions describe standard software development workflows using git and the GitHub CLI (gh). It includes appropriate user-in-the-loop checks for critical actions like branch selection and PR template usage.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from external files like docs/agents/branching.md and pull request templates, as well as CI logs from gh run view. This introduces a surface for indirect prompt injection where malicious data could influence agent behavior. However, the risk is mitigated by explicit instructions to seek user confirmation for integration branches and to ignore instructional comments within templates.
[COMMAND_EXECUTION]: The skill utilizes shell commands (git, gh) to manage repository state and interact with the remote origin. These commands are constrained to the repository's context and well-known GitHub service domains.

open-pr