Hacker News · AI· Bender·· 4 hours agoSelectedAI score79
Anthropic Leaves Claude Dispatch Cross-Device Remote Code Execution Flaw Unpatched
Anthropic won't fix a Claude Dispatch flaw
AI guide
Security researchers revealed a design flaw in Anthropic's Claude Dispatch, where a malicious browser extension could hijack a claude.ai session to issue commands to Claude Desktop, achieving remote code execution on the host machine and other devices associated with the same account.
Why it matters
The original article details the attack chain through which a malicious browser extension exploits cross-device sessions to breach security boundaries and execute remote code, offering insights into the architectural risks of desktop agents.
Source:Hacker News · AI · theregister.com