google-ai-overview

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

• Third-party content exposure detected (high risk: 0.85). The skill calls Frevana’s /service/google-ai-overview endpoint and then validates/parses the returned HTTP response body as JSON and prints it to stdout, so any free-text contained in that outsider-authored API response is ingested into the agent’s LLM context via the tool output.