Anthropic released an update to Claude Code on September 17, 2026, introducing parallel agent workflows that enable users to split coding tasks across multiple threads. This is the company's first major update to the tool since the introduction of autopilot mode as the default setting.
The new feature allows users to describe a goal, and a coordinator splits the work across parallel 'threads,' each running as its own cloud session. Progress is trackable in the main chat or per thread, including on mobile devices.
Claude Code's update is built on cloud-based execution and targets developers seeking more efficient and scalable coding workflows. Availability begins with a beta release for select Pro and Max users using cloud sessions, with team and enterprise access planned for later.
"Claude builds a shared memory across threads over time, and a library collects all uploaded files and results," said Matthias Bastian, a contributor at thedecoder. This functionality allows for greater collaboration and tracking of progress across multiple coding tasks.
The announcement follows Anthropic's recent decision to make autopilot mode the default in Claude Code, claiming it outperformed human developers on safety tasks. The company also highlighted the potential for increased token usage and control over token burn.
Anthropic did not specify the exact timeline for local execution, and the source raised questions about how users will manage token costs and control. The company plans to release more details as the beta progresses.
Source: thedecoder
