The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill reads and processes untrusted input from GitHub issue titles and bodies. This presents a surface for potential indirect prompt injection, where a malicious user could craft an issue description designed to manipulate the agent's triaging logic or cause it to post specific comments. However, the provided instructions include clear bounds on the agent's behavior.
[COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) to interact with remote repositories. It explicitly whitelists gh issue list, gh issue view, gh issue edit, and gh issue comment for its operational needs. This usage is consistent with the skill's stated purpose of issue management.

triage-issue