community-launch-runner

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text is ingested via runtime WebFetch of third-party platform/community rules/docs pages (e.g., Product Hunt, Show HN, subreddit rules, directory submission pages) and those fetched pages’ readable text is used to populate the plan and “self-promotion rule as written” fields, creating an indirect prompt-injection surface from outsider-authored content.