raisindb-messaging-agents

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The document describes agent capabilities that can read and modify every user's nodes, initiate messages to users, and route conversation data to external LLM providers while keeping detailed usage/audit records on the agent side — features that enable straightforward, stealthy data exfiltration and covert remote actions if abused.