Bug: /summarize command fails silently and does not free up chat context

Support Bug Reports
,
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

The /summarize command is currently malfunctioning. When executed, the process terminates almost instantly (within 1 to 2 seconds). Previously, this process took at least 10 seconds or more, depending on the context size.

Current Behavior:

The /summarize execution ends in 1-2 seconds.

The chat context remains fully occupied after execution, indicating that no actual summarization or context compression took place.

Steps to Reproduce

Have a chat with a large/full context window.

Run the /summarize command.

Observe the immediate completion and the unaltered, full context window.

Expected Behavior

The command should process the chat history normally (usually taking 10+ seconds for larger contexts).

The chat context should be successfully summarized and freed up upon completion.

Operating System

Windows 10/11

Version Information

Version: 3.22.7 (user setup)
VS Code Extension API: 1.128.0
Commit: 37076c6c3f9e253c0fa2305197e45befd13a2260
Date: 2026-09-24T05:01:24.589Z
Layout: IDE
Build Type: Stable
Release Track: Default
Electron: 42.10.0
Chromium: 148.0.7778.280
Node.js: 24.18.1
V8: 14.8.178.38-electron.0
xterm.js: 6.1.0-beta.291
OS: Windows_NT x64 10.0.26200

Does this stop you from using Cursor

No - Cursor works, but with this issue

13

Hey @M1L70N, thanks for the report!

We were able to reproduce this on our side. /summarize is being rejected right away when the chat uses Grok 4.7 with the 500K context option, so nothing gets compacted. I’ve flagged it to the team.

In the meantime, this should get /summarize working in that chat:

  1. Open the chat whose context is full.
  2. Click the model name under the chat input and open the Grok 4.7 options.
  3. Set Context to 256K (leave the other options as they are).
  4. Type /summarize and press Enter.
  5. Wait for “Chat context summarized” to appear, then check the context usage indicator.

Automatic summarization still kicks in on its own when the context fills up, so the chat won’t get stuck either way. Let me know if the summary goes through with 256K.

14

Perfect, this workaround solves the problem. It’s a solid improvement over my previous method of having the model summarize, generate a .md artifact, and open a new chat. Thanks for your help!