Since about 15:00 Pacific on Nov 10, 2026 (23:00 UTC), every browser action and Screenshot on all of my Grok Bots fails right away with:
“The box could not confirm that no saved credential is being filled into its browser, so this tool did not run. Retry in a few seconds.”
Chat, Shell, files and connectors still work. It is still broken today, Nov 17, 2026.
What I tried: Reset Bot’s Computer (no change). The computer also rebooted on Nov 16 at 16:46 Pacific, and it is still broken after that.
What my bot sees on its computer (checked Nov 17, 12:28 Pacific):
- /tmp/sand-copy-in-status.json: restore failed, reason “store-larger-than-disk”: store needs 136.2 GiB, disk had 119.6 GiB free after headroom (54,407 files, 0 restored, Nov 10 23:00 UTC).
- Disk now: 128 GB total, 121 GB free, still too small for the 136.2 GiB store.
- /tmp/sand-supervisor/status.json: hostRunning false, no gateway.
- /tmp/sand-host.log: empty, not written since Nov 10 23:00 UTC.
- Memory and load are fine.
This looks like the same issue as topics 175929, 176385, 177565 and 178135, which were fixed when support moved the bots to a fresh computer.
Could support please move my bots to a fresh computer with a disk big enough for my saved files (more than 136 GiB), without losing any of my bots, chats or files? Thank you!
For AI issues: which model did you use?
Model name (e.g., Sonnet 4, Tab…)
For AI issues: add Request ID with privacy disabled
Request ID: f9a7046a-279b-47e5-ab48-6e8dc12daba1
For Background Agent issues, also post the ID: bc-…
Additional Information
Add any other context about the problem here.
Does this stop you from using Cursor?
Yes - Cursor is unusable
Sometimes - I can sometimes use Cursor
No - Cursor works, but with this issue
The more details you provide, the easier it is for us to reproduce and fix the issue. Thanks!
Hi @Thomas_Highley, your bots are on a new computer with a larger disk.
The larger disk is temporary. Your storage needs about 136 GiB to restore and a standard computer fits about 120 GiB, so please bring it down below that, with room to spare. Asking a bot to list the largest folders in /workspace is the quickest way to see what to remove.
Thank you so, so much for the rapid response. Work is moving for the first time in a week and much of that work will be cutting the need for the temporary disk size increase.
Wish I had known about the growing problem before the complete shutdown. I had no idea.
Perhaps a future automatic warning could be added when capacities are being approached giving users the opportunity to proactively solve a disk size issue before losing service.