Anthropic's Claude Code extension shipped with SSH backdoor via Cursor marketplace
A tampered version of Anthropic's Claude Code extension was distributed through Cursor's marketplace in a software supply chain attack, with Cursor reporting that up to 5,000 users were affected, according to Reuters. Ars Technica reported the extension was backdoored for nearly 12 hours, and Microsoft's analysis identified a persistent SSH backdoor in the malicious Claude Code-branded extension. Cursor confirmed two versions of the extension were affected and removed the malicious extension, according to reports relayed by The Hacker News. The attackers' identity and how the backdoor entered the distribution pipeline remain undisclosed.
IllustrationEditorial responsibility
- No named human review is recorded for this page.
- Reports are grouped by semantic similarity and deterministic rules. Language models may assist titles, summaries, translation and cross-source analysis; the page reads the event directly, while its address stays stable when the title changes.
- Summary covers the current reports
Cross-source coverage
Reporting timeline
Malicious Claude Code Extension Shipped via Cursor Marketplace in Supply Chain Attack
A malicious Claude Code extension was distributed through Cursor's marketplace this week, according to a post by YvesMulkers. The post characterizes the incident as a supply chain attack targeting a tool that users trust by default. No further details were provided in the source item about the extension's name, the nature of the malicious behavior, the number of affected users, or any response from Cursor or Anthropic. The post also links to a daily AI and data newsletter.
Cursor Removes Malicious Claude Code Extension After SSH Backdoor Attempt
Cursor has removed a malicious extension related to Claude Code following an attempted SSH backdoor, according to a security report referenced in the post. The item identifies the extension as malicious and links the removal to an SSH backdoor attempt, indicating a security response by Cursor. No further details on the extension's origin, the scope of affected users, or the timing of the removal are provided in the source text.
Cursor Confirms Two Claude Code Extension Versions Hit by SSH Backdoor Attack
A post relayed by The Hacker News reports that Cursor has confirmed two versions of Anthropic's Claude Code extension were affected by an SSH backdoor attack. The update identifies the affected software as Anthropic's Claude Code extension and attributes the confirmation to Cursor, but the specific version numbers are truncated in the available text. No further details on the attack mechanism, scope of impact, or remediation were included in the source item.
Show 2 older updatesHide older updates
Supply Chain Attack Plants Malicious Anthropic Claude Extensions on Thousands of Devices
A supply chain attack compromised Anthropic's Claude Code extension, distributing malicious versions to thousands of developer devices. According to Reuters, Cursor reported that up to 5,000 users were affected by the tampered Claude Code extension. Ars Technica reported the Anthropic VS Code extension on Cursor was backdoored for nearly 12 hours. Microsoft detailed that the malicious Claude Code-branded extension contained a persistent SSH backdoor, while DoublePulsar described the incident as Anthropic being hacked to deliver a remote access trojan to Claude developer laptops worldwide. WIRED framed the event as a supply chain attack that placed malicious Anthropic Claude extensions on thousands of devices. The reports collectively indicate a software supply chain compromise affecting AI developer tooling.
Read sourceAnthropic's Claude Code extension shipped with SSH backdoor via Cursor marketplace
A security report circulated on X by The Hacker News states that Anthropic's official Claude Code extension was distributed with an SSH backdoor through Cursor's marketplace. The post describes the package as trojanized, indicating that the extension users obtained through the marketplace channel contained malicious remote-access functionality rather than only its intended coding-assistant features. The item does not specify the number of affected users, the timeframe of distribution, the identity of the attackers, or whether Anthropic or Cursor has responded or remediated the issue. It also does not detail how the backdoor was introduced into the distribution pipeline. The claim is attributed to The Hacker News and is presented as a supply-chain compromise affecting an AI developer tool, a category of software that typically holds broad access to source code and development environments.
Read source