The project is being developed in public ahead of a planned 1.0 release later in 2026.
OCE adds multi-tenancy, lifecycle governance, auditability, fine-grained permissions and sandboxing. OpenClaw says the system separates trusted from untrusted workloads and allows organizations to swap the harness, model and sandbox for third-party or internal implementations. It can be self-hosted now: the published guidance uses Docker Compose for local development and Kubernetes for internal deployment.
The project originated at OpenAI and was donated to the OpenClaw Foundation. OpenClaw says Red Hat and NVIDIA have collaborated on its development, and that Red Hat and OpenAI are piloting it internally. OCE is free to use, according to the announcement, but is currently intended for internal pilot workloads.
The announcement does not establish production readiness, independent security validation, performance characteristics, or a release date beyond “later this year” for version 1.0.