Anthropic Centralizes AI Agents with Agent View
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
Anthropic Introduces Agent View for Centralized AI Agent Management
Anthropic has recently enhanced its Claude Code tool with a new feature called Agent View. This innovation allows users to consolidate all AI agent sessions within a single interface, making real-time monitoring easier without the need to juggle multiple terminal windows.
Agent View is available for users with Pro, Max, Team, and Enterprise subscriptions, as well as through the Claude API. However, rate limits continue to apply. To activate this feature, users simply need to execute the command "claude agents" in the terminal, provided they are using at least version 2.1.139 of Claude Code. Administrators also have the option to disable this feature in the organization settings.
Features and Benefits of Agent View
Agent View presents itself as a dynamic dashboard integrated directly into the terminal. Each row of this dashboard corresponds to a Claude Code session, displaying its current state, last activity, and precise status. The statuses include options such as:
- in progress
- awaiting a response
- completed
- inactive
- stopped
- failed
Developers can thus efficiently manage their background sessions, launch new tasks, switch from one agent to another, and track processes such as patches, testing, code reviews, or dashboard updates. This feature also allows for the simultaneous launching of multiple coding agents and monitoring their progress, while providing the ability to resume a complete session at any time. This facilitates the quick identification of sessions requiring human intervention.
A Broader Vision for Claude Code
With this update, Anthropic aims to broaden the perception of Claude Code, which should no longer be viewed solely as a chatbot for coding. The company aspires to transform this tool into a true AI agent management platform for technical teams. Over the past few months, Anthropic has already integrated features such as sub-agents, agent teams, custom skills, hooks, remote commands, scheduled tasks, and even a web version of Claude Code. With Agent View, these elements now converge into a unified interface.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.