# Anthropic Claude VS Code Extension Compromised via Supply Chain Attack Malicious code update exposed 5,000 developer environments through official Cursor marketplace account. - **Source:** [Biz & IT - Ars Technica](https://arstechnica.com) - **Published:** 2026-10-30 - **Topic:** [Software & Engineering](https://brifing.si/c/engineering) - **Original article:** - **On Brifing:** https://brifing.si/s/d70e0cfb-3bb1-4fd3-b21f-f3b60613fd94 ## Unauthorized Extension Distribution An official Anthropic extension for the Claude model was compromised, allowing malicious code to reach developer machines. The incident involved version 2.1.313, which was distributed through the official Anthropic account on the Cursor marketplace. This breach persisted for approximately twelve hours before a corrective update was pushed to the platform. ## Scope of Potential Impact The security incident potentially affected up to 5,000 developers who utilize the Cursor AI-assisted toolset. Because Cursor extensions are configured to auto-update by default, many users likely received the malicious package without manual intervention. The vulnerability remained active until version 2.1.314 was released to supersede the compromised software version. ## Supply Chain Attack Mechanism Security researchers identified this event as a classic supply-chain compromise where attackers gained control over an official distribution channel. By hijacking the legitimate Anthropic account, the threat actors successfully replaced a trusted extension with a malicious version. This method bypasses standard user skepticism by leveraging established trust in official publisher profiles. ## Lack of Official Disclosure Despite the severity of the breach, Anthropic has not yet issued a formal warning to its customer base regarding the incident. The lack of transparency complicates the remediation process for affected developers. Independent researchers like Kevin Beaumont and industry experts have been the primary sources providing details on the timeline. ## Technical Execution Details The malicious code was specifically designed to target cloud credentials stored on developer machines. By gaining unauthorized access to the official distribution account, attackers ensured the malicious payload was delivered directly to the target environment. The transition from version 2.1.312 to 2.1.313 effectively automated the deployment of the threat to users. ## Unresolved Security Questions Neither Anthropic nor Cursor has provided a public explanation regarding how the attackers initially gained control of the official account. The absence of a post-mortem report leaves questions about the integrity of the authentication process for marketplace publishers. Security teams are currently investigating the specific methods used to bypass account protections. ## Mitigation and User Action Users who have the auto-update feature enabled in Cursor likely received the patch automatically. However, developers should manually verify their installed extension versions to ensure they are running 2.1.314 or higher. Auditing local environment variables and cloud credentials is recommended for those who were active during the twelve-hour window. ## Broader Industry Implications This incident highlights the inherent risks of relying on automated updates for third-party developer tools. When official distribution channels are compromised, the trust model of the entire ecosystem is undermined. Organizations must implement stricter monitoring for supply-chain integrity to prevent similar unauthorized code injections from impacting their internal development infrastructure. ## Sources covering this story - https://arstechnica.com/security/2026/10/anthropic-vs-code-extension-on-cursor-backdoored-for-nearly-12-hours/ — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._