tech · wired.com

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

A Supply Chain Attack Put Malicious Anthropic Claude Extensions on Thousands of Devices
Image via wired.com
What happened

An unknown threat actor published malicious Anthropic Claude extensions that were automatically installed on roughly 5,000 developer machines. The flawed code in one of the extensions unintentionally blocked the attacker from fully worming the devices, and the breach was publicly disclosed on October 15, 2026, prompting immediate emergency patches.

Why it matters

The incident demonstrates how AI‑powered developer tools can become a new vector for supply‑chain attacks, threatening code integrity and intellectual property across the software industry. It also forces enterprises to reconsider trust models for third‑party extensions, as a single compromised AI plugin can expose thousands of engineers to credential theft and data leakage.

Your next move — for a otherProfessionals across tech should enforce strict vetting of any third‑party AI plugin, implement continuous monitoring for anomalous behavior, and adopt a zero‑trust approach to extension deployment to mitigate similar supply‑chain threats.
Read full article ↗