harness-engineering-orchestrator

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly runs a Market Research agent (SKILL.md § Phase 1) and instructs the Frontend Designer to "research that product's design patterns" when a user provides a reference app/website (agents/frontend-designer.md), which requires fetching and interpreting open/public third‑party websites and references whose content can change planning, stack, and implementation decisions—so it clearly ingests untrusted third‑party content that can influence agent actions.