The Agent Skills Directory

[SAFE]: The skill is a comprehensive reference guide for the official GitHub CLI tool. It describes functionalities inherent to the software such as extension management and shell aliases without including executable code or malicious instructions.
[SAFE]: The documentation describes a surface for indirect prompt injection when the agent processes untrusted GitHub data. Ingestion points: Commands like gh issue view, gh pr view, and gh run view --log read potentially attacker-controlled content from GitHub. Boundary markers: None identified in the provided reference templates. Capability inventory: The tool facilitates shell execution via aliases, file system writes via cloning and downloads, and network access via the GitHub API. Sanitization: No specific sanitization or escaping mechanisms are described for the content being processed.

gh