Issues using Cursor Tab in Jupyter Notebooks

Support Bug Reports
, ,
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Cursor Tab does not work in existing jupyter notebooks / sufficiently old jupyter notebooks. When opening a juypter notebook I worked on previously, there are no Cursor Tab suggestions. If I create a brand new notebook, the suggestions appear as normal. For me, this only occurs with ipynb files and works normally in .py and .html. If I copy cells one by one or tell an agent to copy code to a new file, eventually the suggested line completions stop appearing. It doesn’t appear to be triggered by any specific event. In some cases it gives me garbage suggestions like replacing half-lines I’ve written with comments in prior cells.

I have tried reinstalling / disabling every extension I have and reinstalling cursor itself. I cannot take .ipynb files I made in other ides to cursor and see autocomplete suggestions.

In the dev tools console, there’s a few errors upon launching both new and old files, but the two I think are relevant are: [AiCodeTracking] Error scoring RangeError: stdout maxBuffer length exceeded, and ERR [Extension Host] Error: Start of the range is outside the file. I fed this into a few LLMs and nothing helpful came out though.

Steps to Reproduce

Take an “old” Jupyter notebook and attempt to write new code. I am specifically on Windows 11, running the newest version of Cursor and Jupyter-related extensions as of 6 pm CST 1/11/26.

Expected Behavior

Correct Cursor Tab suggestions

Screenshots / Screen Recordings

Screenshot 2026-01-11 181352.png
Screenshot 2026-01-11 181352.png1019Ă—1019 100 KB

Operating System

Windows 10/11

Current Cursor Version (Menu → About Cursor → Copy)

Version: 2.3.34 (user setup)
VSCode Version: 1.105.1
Commit: 643ba67cd252e2888e296dd0cf34a0c5d7625b90
Date: 2026-01-10T21:17:10.428Z
Electron: 37.7.0
Chromium: 138.0.7204.251
Node.js: 22.20.0
V8: 13.8.258.32-electron.0
OS: Windows_NT x64 10.0.26100

For AI issues: which model did you use?

Tab. Neither auto/fast worked.

Does this stop you from using Cursor

Sometimes - I can sometimes use Cursor

1 Like
3

Hey, thanks for the report.

You’re on a version after the recent fix (Jan 8), so this looks like a new issue. I’ll pass it to the team.

Could you also send the Request ID from the problematic request? Output panel > “Cursor Tab” > find the line that says “CPP Request Log with id”.

4

324b3bb4-01cd-40dc-ba3f-9818e64565da

7

We checked the logs with your Request ID and didn’t see anything obvious. We also can’t reproduce the issue on our side. There were also some recent reliability issues with the provider, can you try again now?

If the issue is still there, can you share a couple details:

  • The size of the old .ipynb files where Cursor Tab doesn’t work
  • About how many cells and lines of code are in those files
  • If you create a new file and start copying cells from the old one, at what point do suggestions stop showing up?

The “stdout maxBuffer length exceeded” error can be related to large files or a lot of output in cells. Please also try clearing output in all cells (Cell > All Output > Clear) and see if anything changes.

8

File sizes are around 100 kb, 20-40 lines of code, ~10 cells.

I just tried copying cells from a broken notebook and I was able to copy the whole thing and cursor tab still worked. However, in a different ipynb file I was testing yesterday that only has 2 cells and 5 lines of code, cursor tab stopped working.

I’ve tried reinstalling cursor and using versions 2.0, 2.1, 2.2, 2.3, but have the same issues on all of them. I tried using different python language servers, disabling cursorpyright, checking their output logs, nothing fixed it.

Given that nobody else is having this issue and you guys can’t replicate it I think it might just be an issue on my end, not with Cursor. Thanks for the help though.

1 Like
10

omg, i ran into this same issue yesterday and am still searching for a solution. i ran the diagnostic tests in the network settings as well and they all pass. reading this, i can confirm that the auto suggestions were working correctly earlier in the old ipynb notebooks but now they seem to be working in all other files except this old notebook i am working on.

11

Hey @wwood and @tabbycat

Haven’t been able to reproduce this, but it would be interesting if you could try the following:

  1. Reload the window (Cmd+Shift+P → “Developer: Reload Window”)
  2. If that doesn’t work, copy the content to a new notebook and see if that changes the behavior
  3. Open up the notebook in a text editor and take a look at the metadata at the bottom
  "metadata": {
    "kernelspec": {
      "display_name": ".venv",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.13.6"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}

Any differences between a working notebook and a not working notebook? Feel free to share them here.

12

it works in a new notebook as well as a new text file in the same directory but doesn’t work for any of the old files (including text). here is the metadata json from one of the old notebooks:

"metadata": {

     "kernelspec": {

      "display_name": ".venv",

      "language": "python",

      "name": "python3"

     },

     "language_info": {

      "codemirror_mode": {

       "name": "ipython",

       "version": 3

      },

      "file_extension": ".py",

      "mimetype": "text/x-python",

      "name": "python",

      "nbconvert_exporter": "python",

      "pygments_lexer": "ipython3",

      "version": "3.12.4"

     }

    },

    "nbformat": 4,

    "nbformat_minor": 5

   }
13

One difference between yours and mine is that mine uses python 3.13.6, and yours 3.12.4. Could you try updating it to see if it makes a difference? It would be surprising, but stranger things have happened!

14

i don’t think python version is the issue because autocomplete works without fail in google’s antigravity on the same files. the way to reproduce this behavior is to create a new ipynb (autocomplete works for new ipynb), then save and close and re open the next day. i created a new file yesterday and it was working well but today when i reopened that file to resume work, the autocomplete doesn’t work in that file. then i created a new file and autocomplete works there.