Urgent.News

What's breaking now, across thousands of outlets.

AI

Claude Code now reads AGENTS.md if there is no Claude.md

Article URL: https://code.claude.com/docs/en/changelog Comments URL: https://news.ycombinator.com/item?id=49760187 Points: 305 # Comments: 129

Anthropic has introduced a new feature called "projects" to its Claude Code tool, allowing users to launch multiple related sessions simultaneously. This feature, available from the Claude.ai sidebar, enables Claude to manage multiple tasks, coordinate parallel threads, and assemble the final result. Users can steer progress, even on mobile devices, and the system continues to work after they step away from their computers.

The new projects capability allows Claude Code to spin up each task as its own cloud session thread operating on its own git branch and copy of the relevant repository. A coordinator directs these threads, and if they touch the same code, merge conflicts may arise. Claude Code now offers five ways to run multiple tasks at once, including subagents, Agent View, Agent Teams, Dynamic Workflows, and the newly introduced Projects.

However, projects consume tokens quickly, potentially reaching usage limits faster. Pro and Max subscribers currently have access to the Projects beta, with Anthropic gradually rolling out the feature to other plans. The rollout is still ongoing, and Team and Enterprise plans do not yet have access to the redesigned Projects beta. Anthropic reported an adjusted operating profit in Q2, with revenue reaching $11.5 billion.

Written by urgent.news from The Register Science's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

This story

This is one outlet's version. Read the fullest account.

Read the original at code.claude.com →

More in AI

Anthropic partners with Accenture to embed evaluators within Anthropic, including red teaming models and conducting alignment assessments (Anthropic)

We're partnering with Accenture on independent evaluation of frontier AI. This is an important step toward the commitment, made in our CEO's essay …

  • Anthropic partners with Accenture for AI model evaluations.
  • Embedded evaluators, including red teaming models, will assess alignment.
  • $1 billion investment over five years for safety and accountability.

More from Friday 18 September →