Letta Conversations API

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

• Third-party content exposure detected (high risk: 0.80). The CLI and demo scripts (scripts/conversations_cli.py and scripts/conversations_demo.py) call client.conversations.messages.list and stream messages via client.conversations.messages.create to fetch and display arbitrary conversation messages from the Letta service, which are user-generated/untrusted content that the agent reads and interprets and therefore could carry indirect prompt injections.