google-maps-research

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

• Third-party content exposure detected (high risk: 0.75). Outsider-authored free text can enter the agent’s LLM context via Places API responses (e.g., places-text/local_market_scan.py include places.name, places.displayName, formattedAddress, websiteUri, and currentOpeningHours), which are returned as JSON and then printed/embedded in the agent output.