newsjack-monitor-setup

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

• Third-party content exposure detected (high risk: 0.95). The required runtime workflow includes a “One real run” step where newsjack monitor run <slug> fetches news from configured RSS feeds and other sources, and the agent then uses newsjack-detector to read and write the results into run.md, so outsider-authored free text from public feeds is ingested into the LLM context via the detector’s analysis pipeline.