OpenClaw has launched version 2.0 in the largest update in the history of the open source project. The release cuts complex setup by auto detecting local models and API keys, rebuilds the browser app, and introduces real time multiplayer workspaces. Following a rare 7 week development pause, the updated platform is now live for all users.
According to the official OpenClaw release notes, this single update represents roughly 50% of all pull requests ever merged into the project repository. Over 16,000 pull requests were completed by 933 contributors, with 569 developers participating for the first time. The team previously maintained a rapid schedule of 106 releases across 230 days. But as the contributor base surged, the project took a 2 month pause to overhaul both the codebase foundation and the internal deployment pipeline.
Getting started used to require tedious configuration. The new version fixes that immediately. First time installation now scans the local computer to identify existing Claude or ChatGPT subscriptions, active API keys, and local models. Users can finish configuring preferences simply by talking directly to their agent. The browser interface was also built again from the ground up. It opens straight into an active conversation where users can manage ongoing automations or track live agent actions.
Everyday use ranges from simple background alerts to complex multi step actions. An agent can watch an inbox for school updates and forward critical notices to Telegram. It can also dig up receipts stored in email and answer family questions on iMessage without manual searching. For larger operations, the update adds shared cloud sessions. This lets multiple team members enter an active workspace to collaborate on tasks while keeping all memory and context completely intact.
