list-hygiene-monitor

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime reads ingestion inputs explicitly provided as an “ESP engagement export”/“bounce/complaint report” (plus an optional webhook event log) and treats the exported rows’ text as untrusted data, so an outsider can submit poison directly into those exports that the workflow then parses to build cohort/trend/worklist evidence.