briefing

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly ingests open third-party sources via engine/collector.py and domain source definitions in domains//sources.yaml (RSS/API), stores that output in data/today_context.json, and feeds it into generation/review prompts and keyword-evolution logic—so untrusted web/social content can materially influence model actions and pipeline decisions.