novel-ralph

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

• Third-party content exposure detected (low risk: 0.10). The required runtime workflow ingests outsider-authored free text only via user-supplied “premise/prompt” and then reads/writes it strictly inside the local working/ directory (e.g., working/state.toml, working/premise.md, working/log.md) rather than monitoring any external queue/feed or fetching arbitrary outsider content.