brainstorm

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to use open web searches and the "WebSearch"/"web-research" collaboration tools (see "Phase 0: Context Loading" and the "Collaboration Tools" section) to fetch market/competitor information from public sites, which the agent will read and use to drive recommendations and decisions, exposing it to untrusted third‑party content.