chatgpt-app-builder

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

• Third-party content exposure detected (high risk: 0.80). The skill's docs and examples (e.g., fetch-and-render-data.md, oauth.md, run-locally.md and CSP/open-external-links sections) explicitly show widgets/tools fetching from external APIs and arbitrary URLs (fetchFlights/fetchOrders, JWKS/userinfo endpoints, ngrok/public URLs, openExternal redirects) and returning structuredContent/content that the LLM reads, so the agent will ingest untrusted third-party content as part of its workflow.