polis-protocol-a-self-optimizing-city-of-agents

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow instructs agents to read their inboxes (_polis/citizens/<self>/inbox.md), which collect messages from other (potentially external or untrusted multi-vendor) agents, representing an indirect prompt injection risk.