z-ai-tools

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

• Third-party content exposure detected (high risk: 0.90). The skill's web_search and web_reader tools explicitly fetch and parse arbitrary URLs (and web pages), and its vision/layout_parsing OCR and GitHub read/search features accept external file or repo URLs, so the agent will consume untrusted public web and user-generated content (e.g., arbitrary websites and GitHub) as part of its workflow.