Cursor AI editor lets repos “autorun” malicious code on devices

Cursor AI editor lets repos “autorun” malicious code on devices

A security flaw in the Cursor AI-powered IDE can lead to automatic execution of malicious tasks when a developer opens infected repositories. Threat actors could exploit this vulnerability to steal credentials, drop malware, or hijack developer environments. #Cursor #WorkspaceTrust

Keypoints

  • The vulnerability arises from Cursor disabling the Workspace Trust feature from VS Code.
  • Opening a repository with a malicious tasks.json file can trigger arbitrary code execution.
  • Threat actors can potentially steal sensitive data or establish C2 infrastructure through this flaw.
  • Cursor developers intend to keep the auto-run behavior despite the security risks.
  • Oasis Security recommends using safer editors for unknown repositories and enabling Workspace Trust.

Read More: https://www.bleepingcomputer.com/news/security/cursor-ai-editor-lets-repos-autorun-malicious-code-on-devices/