frontend

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

• Third-party content exposure detected (high risk: 1.00). The skill's workflows explicitly instruct the agent to fetch and inspect external public resources—e.g., design-consultation.md Phase 2 directs using "WebSearch" and the browse tool to study competing products, design-critique.md says to "pull the design using the Figma MCP tool" when a Figma URL is provided, and design-review.md requires auditing a target URL and extracting page data/screenshots—so the agent will read untrusted third-party webpages/user-provided URLs and act on that content.