Skip to content
NewA calmer, faster ZeroHour

Anthropic VS Code extension on Cursor backdoored for nearly 12 hours

Original title: Anthropic VS code extension on Cursor backdoored for nearly 12 hours

Ars Technica · Securityby Dan Goodin ()1 min read1
Part of a story covered by 12 sourcesBackdoored Claude Code 2.1.313 reached up to 5,000 Cursor usersMerged summary & timeline
AI summary · x-ai/grok-4.7Summaries by Vani AI

Version 2.1.313, published on Anthropic’s official Cursor account, tried to install a persistent SSH backdoor on up to 5,000 machines.

A malicious build of Anthropic’s official Claude VS Code extension, version 2.1.313, was published through Anthropic’s account on the Cursor marketplace and superseded 2.1.312. Cursor said the package tried to add an attacker-controlled SSH key, enable local SSH, and open a reverse tunnel on Windows, macOS, and Linux. Anthropic later published 2.1.314. The report says Anthropic has not publicly explained the account compromise or warned customers.

Unconfirmed: Neither company has explained how the official account was taken over, and the 5,000-user figure is an upper bound.

Key facts
  • Cursor said at most 5,000 users downloaded or auto-installed version 2.1.313.
  • Cursor extensions auto-update by default unless users change that setting.
  • The backdoor code was written for Windows, macOS, and Linux.
  • IDE permissions would expose local files, source code, and developer tools.
  • Kevin Beaumont said Anthropic published 2.1.314 several hours later.
Security assessment · AI, from the article

Impact: Persistent SSH access to developer files, source code, and tooling

ProductAffectedFixed in
Anthropic Claude VS Code extension2.1.3132.1.314
What to do
  1. 1.Replace extension version 2.1.313 with 2.1.314.
  2. 2.Check hosts for unexpected SSH keys and reverse SSH tunnels.
  3. 3.Review Cursor’s extension auto-update setting.
Analyze this story with Vani AIImpact, who is affected and what defenders should do, in a new chat

From the original article

1 min read at arstechnica.com

It's still unknown how attackers got access to Anthropic's verified Cursor account.

Ars Technica · Security

By Dan Goodin · arstechnica.com

Read the full article at Ars Technica · Security

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.