We experienced an error while running the workflow. Sorry about that! You can retry your request, or contact us through our help center at help.openai.com if you keep seeing this error. (Please include the request ID 51cf7be2-3822-47d5-b234-f64ffd70a212 in your message.)
welp, i think you might have to visit that page and send them a message with id you got. might be best to visit another day or in a few hours to see if its been fixed. new releases usually come with some bugs and overtime they get fixed.. think of it like cyberpunk 2077, first day it was full of bugs and now a days there aren’t many… sf most people went home, london everyone is sleeping so its just japan online right now, they are probably looking into it but I’d still reach out to support depending on your urgency
I had same issue today. Try to switch your model for Agent from GPT 5 to GPT 5 mini or GPT 4.1 mini, it helped to me. Also you can try to verify your organisation - I think it should fix it.
I am facing the same issue
Org is verified and tried using GPT 5 mini, GPT 5, GPT
We experienced an error while running the workflow. Sorry about that! You can retry your request, or contact us through our help center at help.openai.com if you keep seeing this error. (Please include the request ID 1aad1229-7bf3-4695-9fc1-4de8c0d54506 in your message.)
has connecting with MCP server worked for anyone?
I have seen a few reports about issues when using while-loops inside the agent-builder.
Are your reports maybe related to this issue?
Same, I suspect that this is related to an issue they are facing with connectors in General, and for some reason, only one MCP server I checked worked, and the others don’t , the one I checked is Context7 → it is stateless and thus I think this is the issue, most MCP servers are statefull, I hosted my own MCP server and looked at the logs, this is what I saw :
Establishing new SSE stream for session 969f6da0-8c1f-4bc5-8eaf-02d0be545034
received delete request
received delete request for session 969f6da0-8c1f-4bc5-8eaf-02d0be545034
So I think :
This is why it worked with Context7 - since it’s stateless…
Hope that they will fix it very soon, cause it is not operational at all at the moment
Hey there!
I were getting the same issue today with my agent, and I don’t think there’s one specific error causing this. OpenAI shows a generic “We experienced an error while running the workflow. Sorry about that! You can retry your request, or contact us through our help center at help.openai.com….“
To find out what the error actually is, I opened up the Agent settings, opened More and under Advanced settings, turned on the Continue on error. Then, you will have Success and Error flows, and on error, you can ask a different, simple agent like GPT-4.1-mini to display what the result of the previous agent was (don’t forget to add the result of previous agent as a context), and next time you run you workflow in preview, when your agent fails, the next agent will let you know why.
This way I found out that the name of mcp server cannot start with a number, and that there was an issue with the response format that I set to the agent.
Hope this helps!
I just noticed after getting the error ALL DAY LONG even on the simplest action, I had to add money to my billing account. I added $5 and then everything started working.
In my case, the issue was with the negative account balance. Adding $10 resolved the issue
I am getting this issue now.. I looked into account balance and changing models also still getting the issue.
My Preview button is not working. I’ve passed the verification, but it’s still dimmed and requesting verification.
I encountered the same error with a guardrail moderate filter on. when turn it off the workflow worked again.
Error: Workflow failed: We experienced an error while running the workflow. Sorry about that! You can retry your request, or contact us through our help center at help.openai.com if you keep seeing this error. (Please include the request ID [ID_HERE] in your message.). (code: unknown_error)