The Agent Skills Directory

[SAFE]: The skill serves as a wrapper for the official GitHub CLI (gh). It facilitates standard development operations such as listing pull requests, checking CI status, and querying the GitHub API. All identified operations are legitimate and align with the stated purpose of the skill.\n- [SAFE]: The installation configuration in the YAML frontmatter uses official system package managers (brew and apt) to install the legitimate gh binary from trusted sources.\n- [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes untrusted data from GitHub. Detection details:\n
Ingestion points: Untrusted content enters the context via gh pr view, gh issue list, and CI logs. (File: SKILL.md)\n
Boundary markers: Absent. The skill does not instruct the agent to use delimiters or ignore instructions within external data.\n
Capability inventory: The skill allows shell execution of gh commands and GitHub API interactions. (File: SKILL.md)\n
Sanitization: No explicit sanitization or filtering of external content is performed. This is a low-risk inherent property of data-retrieval skills.

github