chat-sdk

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

• Third-party content exposure detected (high risk: 0.95). Yes—at runtime the agent receives outsider-authored free text (e.g., message.text / Slack message content) via event handlers like bot.onNewMention(..., message) and passes it into the LLM as agent.stream({ prompt: message.text }) in the required workflow.