Skip to content
NewA calmer, faster ZeroHour

Malicious Claude Code extension reached up to 5,000 Cursor users

Original title: A Supply Chain Attack Put Malicious Anthropic Claude Extensions on Thousands of Devices

WIRED · Securityby Lily Hay Newman ()2 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

A marketplace backdoor stayed live about 11 hours, but a design flaw stopped it on most devices.

A tampered Anthropic Claude Code extension was available in Cursor’s IDE marketplace for Windows, macOS, and Linux for roughly 11 hours on Thursday. Cursor says at most 5,000 users downloaded it and published remediation steps after briefly closing the marketplace. The extension contained a backdoor, but a flaw prevented it from fully establishing itself on most targets. Anthropic and Cursor say Anthropic’s systems were not compromised.

Unconfirmed: How the malicious extension entered Cursor’s marketplace, and whether Cursor itself was breached, is unconfirmed.

Key facts
  • Phorion first spotted the extension on a customer’s Mac.
  • Cursor says no other marketplace extensions were affected.
  • Anthropic says official-channel versions were not tampered with.
  • Cursor has not explained how the upload was authorized.
Security assessment · AI, from the article

Impact: Backdoor access to developer devices running the extension

ProductAffected
Cursor Claude Code extension—
What to do
  1. 1.Follow Cursor’s Thursday remediation steps if the extension was installed.
  2. 2.Use Claude Code only from Anthropic’s official channels.
Analyze this story with Vani AIImpact, who is affected and what defenders should do, in a new chat

From the original article

2 min read at wired.com

Thanks to a lucky break, flawed code in a tainted Claude extension prevented the attacker from fully worming their way into some 5,000 developer machines.

WIRED · Security

By Lily Hay Newman · wired.com

Read the full article at WIRED · 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.