think

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs running "gh issue view {番号} --json title,body,labels" and treating the retrieved GitHub issue text as the initial requirements, so it ingests untrusted, user-generated content from public GitHub issues that can directly influence the agent's decisions and actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime step uses "gh issue view {番号}" to fetch a GitHub issue's title/body and uses that content as the "やりたいこと" prompt (i.e., external content directly controls the agent), so the GitHub Issues API endpoint (https://api.github.com/repos/{owner}/{repo}/issues/{number}) is a required runtime dependency.