The Agent Skills Directory

[SAFE]: The skill performs legitimate GitHub issue triage operations (listing, labeling, and commenting) using the official gh CLI tool.
[SAFE]: Explicit security rules are included to prevent the agent from accessing sensitive information (environment variables, secrets) or executing unauthorized administrative commands such as gh auth or gh secret.
[PROMPT_INJECTION]: While the skill processes external data from GitHub issue titles and bodies, it does not display patterns of direct prompt injection. The potential risk for indirect prompt injection is significantly mitigated by the narrow scope of permitted tools and the explicit constraints on permitted actions.

triage-issue