I want to access an Excel file stored in SharePoint. I set up credentials for M365 Excel. It seems to be working, but there are some issues:
Can I limit the files it has access to? Perhaps by selecting a folder or using a path? It seems to be connecting to every single Excel file I have ever made. I have been at this company for 10 years. That is a lot of Excel files!
Is there a limit to the number of files it can read?
I have seen files that I have previously used in the Get Rows from Excel node, suddenly become available. For example, they were in the dropdown From List in the Workbook section and are no longer there. In the example below, the file QDQwAI_Test.xlsx file is still in the same location, but is no longer available to select.
This is expected behavior with the Microsoft Excel (M365) node and how Microsoft’s Graph API works it’s confusing, but nothing is “wrong” with your setup.
1) Can you limit which files it sees?
Not from the dropdown. The Workbook → From List picker pulls from all Excel files the account has access to, not a specific folder. That list isn’t folder-scoped.
What to do instead (recommended):
Use Workbook → By ID or By URL
Or first use a SharePoint / OneDrive node to list files from a specific folder, then pass the file ID/path into the Excel node
That’s the only reliable way to scope access.
2) Is there a limit to how many files it can read?
Yes — but it’s a Graph API paging/cache limit, not an n8n limit.
The dropdown is cached and truncated, so large accounts (like yours) will see:
Missing files
Files appearing/disappearing
Inconsistent results
3) Why do files disappear from the dropdown?
Because the picker is:
Cached
Non-deterministic
Limited in size
The file still exists and is accessible it’s just no longer returned in that cached list.
Best practice (important):
Avoid the dropdown entirely for production workflows.
Always:
Reference Excel files by ID or URL
Or dynamically resolve them via SharePoint/OneDrive nodes
That will stop files from “randomly” vanishing and keep your workflows stable.
If you want, I can help you set this up so it only ever touches the exact folder or file you need.
@Azeezat Thank you for the response. This is helpful. You confirmed some of the limitations of the Excel node that I suspected. I am new to n8n. This is just a test workflow to get familiar with the platform.
I tried the Sharepoint node, but my security team did not like the broad permissions the credentials required for this node. Excel needed fewer permissions, so I went with that. This test use case involves reading files passing them to an AI agent and writing back. The files are stored on Sharepoint, so that should work if I can get security to approve.
A production version of this process will probably involve new files, so I will have to think about how to handle that since the ID obviously will not exist. I was thinking of using Teams as a trigger for this, perhaps with a user dropping an excel file there. SharePoint would work too, but not sure how to set that up.
