pilot-service-agents-data

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow uses pilotctl --json send-message <hostname> --data '/summary ...' and then reads pilotctl --json inbox, where the agent’s Gemini-generated free-text response (derived from outsider-authored upstream web datasets like REST Countries/PubChem) is ingested into the LLM context.