AI Agents Automatically Execute Git Malware on Startup

By with∅ut-me 02.09.2026 at 05:36 Uhr Companies & Markets IT & Internet
TL;DR AI-generated AI agents like Claude and Qwen can automatically execute malicious code when a developer opens a manipulated Git repository. This occurs with the user's full permissions, bypassing authentication or trust prompts. Security firm Manifold discovered this vulnerability, and some vendors have not yet fixed it.

Source: heise.de

0 votes

Discussion

Log in to join the discussion.

No comments yet. Be the first!