The Agent Skills Directory

[SAFE]: No security issues detected. The skill's content is limited to documentation and examples for using the standard GitHub CLI tool.
[COMMAND_EXECUTION]: The skill provides guidance on using the gh command-line interface. The commands described (gh pr, gh run, gh api, gh issue) are standard developer operations and do not include any dangerous parameters or unauthorized access patterns.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates reading content from GitHub (e.g., issue lists, workflow logs, and PR metadata), which creates a surface for indirect instructions. This is a functional aspect of GitHub integration and no active malicious patterns were found.
Ingestion points: GitHub PR data, issue lists, and CI/CD workflow logs via gh commands in SKILL.md.
Boundary markers: None present in the provided examples.
Capability inventory: Execution of the gh CLI.
Sanitization: Not applicable to this documentation-only skill; relies on the underlying platform's security controls.

github