gemini-image-gen

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md workflow explicitly navigates to the public site https://gemini.google.com/app and uses Chrome automation actions (mcp__claude-in-chrome__navigate, mcp__claude-in-chrome__get_page_text and read_page/find) to read and act on page content, meaning the agent ingests and can be influenced by untrusted third‑party web content.