The new Folder REST APIs doesn't work

Hi ​@mobile_dev, thanks for reaching out! 

I understand how this must be disrupting to your workflow. The endpoints and scope you're using are correct, so this shouldn't be happening. One detail that may be helpful: a scope mismatch would return a 403, not a 404. Since the same team ID works on /v1 with your older token, we can rule out both the token scope and the team ID, which points to something on our side rather than your setup.

Two quick things to help us pin it down:

1. What's the full response body on the 404 (the message field)?
2. When you open the Figma file browser, do you see the new Folders experience with nested subfolders yet, or still the older Projects view?

That second one matters because the Folders rollout is still in progress, and it'll tell us whether the endpoints are gated for your org. In the meantime, your existing projects:read token will keep working. All existing Projects endpoints remain supported, and GET /v1/teams/:team_id/projects now returns your top-level folders. I know that's a stopgap and not a fix, since you can't mint a new token with that scope.

Let me know, and I can flag this to the team that owns these endpoints. Here are some docs for reference in the meantime: Folders endpoints and Scopes.

Hi ​@djv 

Thanks for your reply. Here’s the information you asked for:

Full response body for the 404:

{"error":true,"status":404,"message":"Not found","i18n":"error.not_found"}

Folders experience in the UI:
Yes, I do see the new Folders experience in the Figma file browser — there's an "All shared folders" button visible. However, in the Recent view, I still see only the old Projects, not folders. 

Regarding the token: I understand that the old projects:read token continues to work, but the issue is that all of our existing PATs expired just a few days after the Folders update — their 90-day lifetime ran out. Since the UI no longer allows creating new tokens with projects:read, and the new folders:read scoped tokens return 404 on the new endpoints, we're effectively locked out of programmatic access to our team's files.

Alternatively, if the new Folders APIs aren't ready yet, could you please temporarily bring back the ability to create new PATs with the old projects:read scope? That would at least unblock us while the new endpoints are being sorted out.

We'd really appreciate a quick resolution here, as this has completely broken our CI/CD pipeline.

Let me know if you need any additional details.

Thanks for the additional details, ​@mobile_dev! 

I've escalated this to the team that owns these endpoints, including your request to temporarily re-surface projects:read. I can't promise that specific fix or a specific timeline yet, but I'll update this thread as soon as I hear back.

One thing that might unblock CI in the meantime: does your pipeline need to discover files, or does it work against known file keys? If it's the latter, a new PAT with file_content:read / file_metadata:read should work today. Those endpoints are untouched by the Folders migration. Let us know, and we’ll keep you updated when/if we hear back from the team. 

Thanks for the quick response and for escalating this to the team - we really appreciate it.

To answer your question: yes, our pipeline currently needs to discover files. It fetches the list of all projects/folders in the team and then selects the appropriate one dynamically at runtime based on the context. So working with known file_key values isn't an option for us in the normal flow.

That said, since we don't have a clear timeline for the fix, we will implement a temporary workaround on our side - we'll switch to using known file_key values hardcoded in our configuration instead of discovering them automatically. This should unblock us for now, even though it's not ideal and adds manual overhead.

In the meantime, we'll be waiting for the team's decision regarding the new Folders endpoints and/or the temporary return of the projects:read scope for new tokens.

Thanks again for your help!

Hi ​@mobile_dev, thanks for your patience! 

Our team was able to reproduce the issue and just pushed a fix. Can you please refresh your tabs and try again? 

Let us know how it goes!