audit-ui-and-save-files

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow requires invoking /run-agent-browser against a user-supplied base-url (Phase 1 / Phase 3) and may enumerate routes from sitemap.xml and read live pages/screenshots, so it fetches and interprets arbitrary public or user-provided web content which can materially influence clustering and approval-gated fix-dispatch decisions.