The Agent Skills Directory

[NO_CODE]: The skill is composed entirely of markdown documentation and does not include any scripts, binary files, or installation logic.\n- [SAFE]: No malicious patterns, direct prompt injections, or unauthorized network operations were detected. The skill explicitly warns against using commands that modify or delete resources, steering the agent toward safe read-only tasks.\n- [PROMPT_INJECTION]: The skill defines commands to fetch external data (issues, pull requests, logs) from GitHub which are potentially controlled by third parties, creating an indirect prompt injection surface.\n
Ingestion points: gh issue view, gh pr view, gh run view --log, and gh api commands in SKILL.md.\n
Boundary markers: Absent in the documentation.\n
Capability inventory: Interaction with GitHub via the gh CLI tool.\n
Sanitization: Absent; the skill is a reference and does not provide sanitization logic for the command outputs.

gh-reference