Crypto Games ExtensionCrypto Games AppTelegram NewsSubmit
Anthropic relaunches Claude Code Projects to run a team of cloud agentsImage Source: TheVerge

Anthropic relaunches Claude Code Projects to run a team of cloud agents

George Tsagkarakis 2 min read
Contents 5 sections
We may include affiliate links in our content, meaning we could earn a commission—or receive blockchain-based assets—if you click a link and make a purchase or take a specific action. Additionally, we use generative AI to help draft and refine our posts for clarity and grammar. All content is fact-checked and reviewed by a human editor before publication.

Anthropic’s pitch for the rebuilt Projects feature in Claude Code isn’t a better chat window. It’s a team of agents that talk to each other.

Projects now lets you run multiple agents under the same roof, sharing memory, goals and a library of files and artifacts. Each project spins up “threads” that handle different tasks in parallel, with a “coordinator” directing the work. If you’ve used Grok Bot or any of the other tools that herd groups of AI agents, the shape will look familiar.

STKB364_CLAUDE_2_C_96d15c
Anthropic relaunches Claude Code Projects to run a team of cloud agents 1 Anthropic relaunches Claude Code Projects to run a team of cloud agents

Every thread is its own branch

Here’s the part that matters to anyone who’s ever babysat a bot loose in a repo. Under the hood, each thread is a Claude Code cloud session working on its own branch and its own copy of the repo.

The coordinator keeps the work organized. But when two threads touch the same code, Anthropic isn’t promising magic. The overlap surfaces as a merge conflict, same as any other PR you’d resolve by hand.

That’s an honest design choice rather than a fix. Git already knows how to tell you two people edited the same function, and Projects is leaning on that instead of pretending a coordinator can arbitrate intent.

Threads can split themselves further

A thread isn’t the smallest unit here. Each one can break its delegated work into pieces using subagents, loops and workflows, which is how Anthropic says large assignments finish faster.

So the hierarchy runs three deep: project, thread, subagent. Whether that holds up on a repo with real history is the question nobody can answer from a launch post.

The cloud-only catch

At launch, threads run in the cloud. Support for local tools and code is coming “very soon,” according to Anthropic, which is the kind of timeline that means what you think it means.

For anyone whose build depends on a local toolchain, a local database or anything that doesn’t exist inside a cloud sandbox, that’s not a minor asterisk. It’s the difference between using this today and bookmarking it.

Who actually gets it

The updated Projects feature is in beta starting today for “select Claude Pro and Max subscribers.” Not all of them. Select ones.

Access expands later to all Pro, Max, Team and Enterprise users, plus Cowork and regular Claude chats. No date attached to “later.”

If you’re on Pro and you don’t see it, you’re not doing anything wrong. You’re just not in the select group yet, and the most useful thing in this release for most developers, running threads against code on your own machine, hasn’t shipped at all.

Share this article
George Tsagkarakis

George Tsagkarakis, known as Staycalm4now is a professional author in the crypto gaming industry since early 2018. He has experienced all the growth of Blockchain Gaming and helped multiple projects achieve their goals and established a player base. He is the co-founder of egamers.io and now the Founder and owner of CryptoGames.gg He is also the COO of MyStage, an…

More from AI Agents

Leave a Reply