agent-profile

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

• Third-party content exposure detected (high risk: 0.80). The skill fetches and reads public ATProto agent profiles and cognition collections (e.g., via the https://pdsls.dev/at/did:plc:AGENT_DID/network.comind.agent.profile/self endpoint and the tools.registry get/list/query commands), which are untrusted, user-generated third-party content the agent is expected to interpret.