deep-research

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly automates opening a Gemini "Deep Research" browser tab and submitting research queries (see SKILL.md "How It Works" and "Usage"), which implies ingestion and interpretation of open/public web content returned by Gemini's research feature and therefore exposes the agent to untrusted third‑party content that could carry indirect prompt injections.