create-faceless-channel

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

• Third-party content exposure detected (medium risk: 0.30). Outsider-authored text is ingested at runtime via the configurable discovery/capture pipeline that “fetch[es] candidates from a documented API, feed, page … or approved scraper” and then “use[s] a real browser, scan[s] the page” to write scripts from “source-specific facts,” with the source entrypoint supplied in the channel config.