Anthropic: Claude Code gets Agent View for managing parallel sessions

Anthropic shipped Agent View to Claude Code on May 11, 2026 as a Research Preview. The feature introduces a session roster inside the CLI: a single screen where developers can see all active coding sessions, each showing its current status (working, waiting for input, completed, failed, idle, or stopped) along with the time of last activity.

Previously, running multiple Claude Code tasks at once required separate terminal tabs or a multiplexer like tmux. Agent View replaces that with a unified control surface. Developers open it by running claude agents directly in the terminal, or by pressing the left arrow from any existing session. From there, spacebar shows a preview of recent activity, Enter opens a full transcript, and a session can be pushed to the background with /bg or launched there from the start using claude --bg "".

The feature is available on Pro, Max, Team, Enterprise, and Claude API plans, with standard rate limits.

For product managers, the significance is twofold. On engineering teams adopting agentic coding workflows, Agent View signals that AI-assisted development is no longer a single-thread activity — it is shifting toward an operational model where multiple tasks run in parallel and developers step in only when the system needs a decision. PMs who plan sprint capacity or evaluate AI tooling for their teams should understand this change in how developers interact with AI assistants. On a practical level, PMs who use Claude Code themselves to prototype or run data analysis can now run several workstreams without losing track of what each agent is doing.