api-testing

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow drives a browser session to a user-supplied URL (e.g., navigation go-to --url "https://app.example.com"), which can cause the LLM to ingest outsider-authored free text from fetched web page content (public web content) via content get-as-text/take-screenshot and related page-reading steps.