Anthropic has launched Agent View for Claude Code, introducing a centralized interface for managing multiple AI coding agents simultaneously. The feature is available as a research preview for users on Pro, Max, Team, Enterprise, and Claude API plans.
The new interface is designed to simplify parallel AI-assisted development workflows that previously required developers to manage multiple terminal windows, tmux sessions, or separate command-line tabs manually. Agent View allows users to monitor the status of multiple Claude Code sessions from a single screen, showing which agents are actively working, waiting for user input, or have completed their assigned tasks.
Users can open the interface through the claude agents command or by pressing the left arrow key from an active session. Each row in the dashboard displays the current session state, the latest response, and the last interaction timestamp. Developers can preview recent outputs without leaving the main interface and respond inline when a session requires approval or clarification.
Anthropic also introduced background execution support. Existing sessions can be moved into Agent View using the /bgcommand, while entirely new background tasks can be launched directly with claude --bg [task]. The company says the workflow is intended to reduce context switching while enabling developers to coordinate multiple coding agents at once.
Another major addition is a continuous execution mode activated through the /goal command. The feature allows Claude to continue operating without interruption until a defined objective is completed. Anthropic positions the tool as a way to support long-running engineering workflows such as automated pull request reviews, dashboard maintenance, or iterative coding tasks.
Early users have reportedly adopted the system for parallel experimentation, long-running automation jobs, and rapid navigation between coding sessions. Anthropic said developers are using Agent View to dispatch several independent tasks simultaneously and later review completed pull requests from a unified dashboard.
What It Means
The update reflects a broader shift toward multi-agent AI development environments where several AI systems work concurrently under human supervision. Instead of interacting with a single chatbot session, developers increasingly orchestrate multiple specialized agents handling different tasks in parallel.
For software teams, the feature could improve productivity in large projects that involve debugging, code generation, testing, and documentation simultaneously. Background execution and centralized monitoring may also make AI-assisted development more practical for enterprise engineering workflows where interruptions and manual coordination slow progress.
The continuous execution mode is particularly notable because it moves Claude Code closer to semi-autonomous software workflows. Rather than requiring frequent prompts, developers can define an objective and allow the system to continue iterating until completion or intervention becomes necessary.
