cloud-monitoring-chart-generation

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

• Third-party content exposure detected (medium risk: 0.30). Stage 2/3 Runtime path ingests outsider-authored free text from the user prompt and/or query JSON into the LLM and then into scripts/assemble_widget_proto.py via --spec_json, --promql_query, and --lts_request_json where strings are incorporated into the generated Widget textproto.