Skip to main content

Claude Code adds a manager view for parallel agent sessions

By Steven Van ·

The research preview lists every session's status and lets you reply inline, available now on Pro, Max, Team, Enterprise and API plans.

Claude Code has added agent view, a single screen for managing every running session. Press the left arrow from any session, or run claude agents, to see a list showing each session's status, whether it needs input, its last response, and when it was last touched.

  • Select a session to peek at its last turn; if it's waiting on a decision, reply inline and it picks back up. Press enter to attach to the full transcript.
  • Add an existing session to agent view with /bg, or start a new one directly in the background with claude --bg [task].

Anthropic says early users have used it to run several sessions in parallel and return to a list of pull requests ready for review, to track long-running loops like PR babysitters and dashboard updaters (which show their next run time in the list), and to jump between sessions without losing a running task.

Agent view is a research preview available now on Pro, Max, Team, Enterprise, and Claude API plans, requiring Claude Code v2.1.139 or later. Opt in by running claude agents; usual rate limits apply. Read the full announcement.

Claude
Claude
AI assistant by Anthropic — write, analyze, code, and create with one of the most capable large language models available.
View Claude →

Original source

Read Claude Code adds a manager view for parallel agent sessions on Creators Toolbox