github-triage

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow fetches outsider-authored GitHub issue/PR content (including body and comments via gh issue list/gh pr list JSON fields) and passes it into the quick subagent prompt as {body} / {comments_summary}, which becomes readable free text in the agent’s LLM context (indirect prompt injection risk).