AIToday
Large Language ModelsAI Coding AssistantsThe Verge AIPublished: Sep 18, 2026, 04:00 JST

Claude Code beta opens multi-agent projects to select Pro and Max subscribers

Claude Code beta opens multi-agent projects to select Pro and Max subscribers

3 Key Points

  1. What happened

    Anthropic launched a revamped projects feature in Claude Code in beta today for select Claude Pro and Max subscribers, allowing multiple AI agents to run under one roof with shared memory, goals, and files.

  2. Why it matters

    The coordinator and parallel threads let users manage complex coding tasks in one place, with overlaps resolved as merge conflicts—likely reducing the need to switch between separate agent sessions.

  3. What to watch

    Access expands later to all Pro, Max, Team, and Enterprise users, plus Cowork and regular Claude chats, while local tools and code support is promised 'very soon'; the test is whether cloud-only threads meet developers' local workflow needs.

WHO IT HITSDevelopers using Claude Code for parallel coding tasks get a shared-memory, coordinator-led way to run multiple agents. Enterprise teams on Pro, Max, Team, or Enterprise plans will likely see it later, and IT admins may need to plan for cloud-based agent sessions before local tools arrive.

Not sure about something? Ask the AI

Summaries like this, in your inbox every morning.

Context & Analysis

Anthropic is positioning Claude Code projects as a way to corral multiple AI agents inside a single workspace. The design keeps each thread on its own branch and copy of the repo, while a coordinator directs activity and resolves overlapping work as merge conflicts, much like a normal pull request. This mirrors other tools that manage groups of agents, such as Grok Bot.

The launch is deliberately staged: beta today for select Claude Pro and Max subscribers, with broader access later for Pro, Max, Team, and Enterprise users, plus Cowork and regular Claude chats. Initially, threads run in the cloud, though Anthropic says local tools and code support is coming 'very soon.'

For developers juggling several coding tasks, the appeal is less context switching and a shared memory across agents. The outcome likely hinges on how well the coordinator handles overlapping edits and how quickly local support arrives, since many teams still rely on local development environments.

FAQ
Who can use the new Claude Code projects feature?
It is available in beta starting today for 'select Claude Pro and Max subscribers.' Access will expand later to all Pro, Max, Team, and Enterprise users, as well as Cowork and regular Claude chats.
How do parallel agent threads avoid conflicts?
Each thread is a Claude Code cloud session on its own branch and copy of the repo. If threads work on the same code, the overlap is resolved as a merge conflict just like any other PR.
Can threads split work further?
Yes, each thread can split its delegated work into pieces using subagents, loops, and workflows when needed, so large assignments finish faster.

Get the latest Large Language Models news every morning

For example, today's edition would include:

  • Workiva Amplify: traceable AI agents needed for financeSiliconANGLE AI · 1h ago
  • UC Berkeley: harness cuts AI answer cost 71%Tomasz Tunguz (Theory Ventures) · 1h ago
  • Anthropic rebuilds Claude Code Projects for parallel agentsTHE DECODER · 1h ago

AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. Q&As are published on this page for other readers too.

Related Articles

Next articleBaseten's Base Labs, Hugging Face, Goodfire to build open-weight safety standard