Omar Shahine said a preview of Microsoft Autopilot, a persistent personal agent, is reaching Microsoft’s first customers and is built on OpenClaw. Peter Steinberger said work with Microsoft to prepare the codebase for large-scale deployments has been underway since March.
Supporting plugins in OpenClaw 2026.9.6 can use an optional model for structured choices, separate from chat. TypeSafe Jev sends supplied information to its hosted API and incurs normal charges, while ONNX offers local CPU choices. Decision models are off by default.
The 2026.9.6 release also lists managed updates, restart recovery, 30-day usage, a GitHub reader, remote files, memory and skills, live meeting notes, and Jev with decision models. OpenClaw cites 2,614 pull requests and 351 contributors.
OpenClaw’s next version uses a decision model to choose automatically between steer and queue. The lab feature supports Jef*, API-compatible models, including local models such as Kef..., and ONNX variants.
OpenClaw says the support was pushed last week, and describes how it is thinking about using decision models and how people can contribute.
Peter Steinberger quoted Nat Friedman to counter a circulating “Meta uses OpenClaw” story, saying they built their own agent and were inspired by OpenClaw. Friedman said Muse was built from scratch but is heavily inspired by OpenClaw.
josh avant said OpenClaw received a security audit from Trail of Bits via OpenAI’s Patch the Planet initiative. Peter Steinberger quoted that post and said they found nothing critical.
