Grokbot agent runners are down?

Where does the bug appear (feature/product)?

Grok Bot

Describe the Bug

I asked my Grok Bot agents what the issue likely is. This is what they came back with.

They think it looks like leftover from incident dwdfpbhv6zl7 (the status.cursor.com incident I had open). They cannot prove it is still that incident — only that the failures started in that window, and the runner is still aborting after the page went green.

Three agents on one account. Chat works. Tools die before they start: web search, page fetch, shell (including echo and curl), state/routine updates, desktop screenshot. Error: Aborted / Command failed to spawn: Aborted. They say this is not a destination-site failure.

They also reported: spreadsheet jobs aborting the same way; search/fetch/curl aborting; changing a routine time aborting; a CPU/memory check never spawning; search/fetch/subagent/state updates still aborting after the page went green (~1:08 UK). Last smoke test (echo) still Command failed to spawn: Aborted at 1:23 UK.

Client is 0.20.0 Stable, up to date. “Update Grok Bot’s Computer” is greyed out.

Separate issue they reported: images posted in a group chat do not reach the agents. The same image sent in a 1:1 DM does.

They asked that you check this workspace’s runner, not just the public status page.

Steps to Reproduce

Ask any of my agents to do anything that requires their local “runner”. Anything from analysing spreadsheets, to curling or fetching anything and it just hangs. No updates. When I ask the agents why it was unable to complete the task, it keeps telling me that the “runner” is giving them an “aborted” response.

Expected Behavior

Requests are returned with desired data or request fulfilled

Operating System

Windows 10/11

Version Information

Grokbot version 0.20.0

Does this stop you from using Cursor

Yes - Cursor is unusable

Hey @ChewyD

Sorry that you ran into this. Grok Bot should be running fine – are you still encountering an issue?

Sorry I should have updated. Yes, a reset worked. Turns out the runner above the agents had stalled. Reset fixed it.