iterate-post-launch

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

• Third-party content exposure detected (medium risk: 0.30). In required runtime phases, the agent reads free-text/outsider-authored content from Sentry issue titles/details via sentry:search_issues and sentry:analyze_issue_with_seer, which can contain user/third-party stack traces and comments before the agent selects a specific trusted item.