Cloud Agents crash with "We encountered an internal error - please try again in a moment."

Bug Reports
1

Where does the bug appear (feature/product)?

Background Agent (GitHub, Slack, Web, Linear)

Describe the Bug

Our background agents frequently stop executing with the error message ‘We encountered an internal error - please try again in a moment.’ This happens after our Cloud Agents run long sessions that reach very large token amounts (around 500k tokens). When this occurs, the agent stops until we manually resume by opening the Cloud Agent and sending a message such as ‘It looks like you were interrupted. Please continue.’

Steps to Reproduce

  1. Set up a Cloud Agent with a background agent (e.g., GitHub/Slack/Web/Linear).
  2. Run the agent on a task that will cause it to accumulate a very large token count (e.g., around 500k tokens).
  3. After some time, observe that the agent stops executing and displays ‘We encountered an internal error - please try again in a moment.’
  4. To continue, you need to manually open the Cloud Agent and send a message such as ‘It looks like you were interrupted. Please continue.’ This works for a few minutes until the Cloud Agent inevitably crashes again or finishes the task.

Expected Behavior

The Cloud Agent should continue executing background agents without stopping due to internal errors, even when token usage is large. It should not require manual intervention to continue.

Screenshots / Screen Recordings

Screenshot 2025-11-19 at 11.42.43 AM.png
Screenshot 2025-11-19 at 11.42.43 AM.png1882×1404 440 KB

Operating System

MacOS

Current Cursor Version (Menu → About Cursor → Copy)

Version: 2.0.77
VSCode Version: 1.99.3
Commit: ba90f2f88e4911312761abab9492c42442117cf0
Date: 2025-11-13T23:10:43.113Z
Electron: 37.7.0
Chromium: 138.0.7204.251
Node.js: 22.20.0
V8: 13.8.258.32-electron.0
OS: Darwin arm64 23.3.0

For AI issues: which model did you use?

GPT 5 Codex High

For AI issues: add Request ID with privacy disabled

bc-888cb16e-297f-4eb1-81e5-23c6224532d1

Does this stop you from using Cursor

Sometimes - I can sometimes use Cursor

2

Hey, thanks for the report. It looks like this is a real issue: Cloud Agents are hitting resource limits during long-running sessions with high token usage.

I’ll escalate this to the engineering team for investigation.

In the meantime, please keep using your workaround of manually resuming the agent with “Please continue”, although I understand this isn’t ideal for production workflows.

Thanks again for the detailed report with all the necessary information.

1 Like