gemini-json-parsing

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

• Third-party content exposure detected (high risk: 0.80). The skill parses Gemini CLI outputs (e.g., .response and stream-json events like "tool_result") and explicitly references tool outputs such as "google_web_search", so it ingests untrusted third-party/user-generated content returned by the CLI as part of its workflow.