create-a-skill

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (Phase 1 "Research the domain" in SKILL.md) explicitly instructs the agent to "Use web search to look up best practices, official documentation, or reference implementations" and to research via subagents, meaning the agent will fetch and read open/public third‑party content (which can influence decisions); the repo also includes viewer HTML that pulls external CDNs (fonts/SheetJS), further showing runtime exposure to third‑party resources.