responsive-screenshots

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime path is triggered by a user-provided URL (“Responsive screenshots of https://...”), and it uses the Playwright MCP to fetch/render that page and then the agent analyzes the resulting page content/screenshots; this can include outsider-authored free text from the target webpage that the LLM ingests during analysis.