As reported by the official account of the Artificial Intelligence Industry Development Alliance, on April 10, during the 17th General Meeting of the China Artificial Intelligence Industry Development Alliance, its Security Governance Committee officially unveiled the “Guidelines for Risk Management in Deploying Agents Similar to OpenClaw”. In response to the security challenges that agents akin to OpenClaw present, these guidelines put forward a security deployment framework and self-inspection standards that encompass the entire lifecycle, from deployment and usage to decommissioning. They specifically target the pain points in risk prevention and control that enterprises encounter during technological implementation. This move is designed to steer agents from merely being “deployable” to achieving “secure deployment, standardized usage, and ongoing governance”, thereby offering practical guidance for enterprises in their technology deployment endeavors.