research

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow ingests outsider-authored free text from public web sources via the scholar lookup / scholar search CLI and the consensus CLI (their JSON outputs include titles/snippets/journal strings scraped from the public web), which the script then prints into the agent’s LLM context as NDJSON source/final events.