figure-out-team

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

• Third-party content exposure detected (high risk: 0.85). High: the runtime workflow polls a Slack thread/channel for “new messages since your cursor” via the slack-poller subagent, and those outsider-authored Slack message bodies are ingested as readable text into the agent’s LLM context (indirect prompt injection risk).