22 days ago
My workspace (Motyva's Projects, Hobby plan) hit the Agent Usage limit ($5.08/$5.00) this billing cycle (Jul 26–Aug 26), which triggered "Limited Access" and paused deploys on the behavioral-insight service.
I did not use the Agent manually this cycle. Based on your docs (docs.railway.com/ai/railway-agent), I understand the Agent runs "Automatic deployment diagnosis" whenever a deploy fails, with no manual trigger required. We had several failed deploys today (Jul 30) in a row during a round of bug fixes, which would explain the usage.
My questions:
Is there a way to specifically disable "Automatic deployment diagnosis" (auto-diagnosis on failed deploys), while keeping the Agent available for manual use when I want it?
If there's no dashboard toggle for this, can you disable it on your end for my workspace?
Since this usage wasn't an active choice on my part, is it possible to review/refund part of the amount generated by this chain of failed deploys?
I've already raised the limit manually to unblock deploys, but I'd like to avoid this happening again uncontrolled during future rounds of failed deploys.
Thanks, Bruno Amaro Workspace: Motyva's Projects
1 Replies
22 days ago
There is no toggle to disable only automatic deployment diagnosis while keeping manual Agent use. To prevent all Agent charges, set the Agent hard limit to $0 at your workspace usage page via "Update limits" - this disables Agent features entirely while compute and deploys continue normally. We cannot disable it selectively on our end either. Regarding a refund, Agent usage is billed for tokens actually consumed and is a valid charge whether triggered manually or automatically, so we are not able to refund it.
Status changed to Awaiting User Response Railway • 22 days ago
15 days ago
This thread has been marked as solved automatically due to a lack of recent activity. Please re-open this thread or create a new one if you require further assistance. Thank you!
Status changed to Solved Railway • 15 days ago