Cursor processing someone else's context

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

I have been working on my project and getting Agent to resolve my GCP service account issues. Suddenly after one of the messages i saw the agent processing something completely out of context. I suppose that this is other person’s chat. I stopped the execution multiple times and on retry i got different agent logs in multiple languages and various topics. I was using Sonnet 4.5 thinking. When i changed to Opus it got fixed. Now i am unsure whether my information is safe and if nobody else got access to my context.

This is a critical security issue that makes me really concerned.

Steps to Reproduce

Use the agent with Sonnet 4.5

Screenshots / Screen Recordings

Operating System

MacOS

Current Cursor Version (Menu → About Cursor → Copy)

Version: 2.1.47 (Universal)
VSCode Version: 1.105.1
Commit: 2d3ce3499c15efd55b6b8538ea255eb7ba4266b0
Date: 2025-12-04T02:31:50.567Z
Electron: 37.7.0
Chromium: 138.0.7204.251
Node.js: 22.20.0
V8: 13.8.258.32-electron.0
OS: Darwin x64 24.6.0

For AI issues: which model did you use?

Sonnet 4.5 thinking

Does this stop you from using Cursor

No - Cursor works, but with this issue

Do you have a request ID where this happens?

This topic was automatically closed 22 days after the last reply. New replies are no longer allowed.