Error: 400 Your request cannot be processed due to a validation error

Questions
,
1

Hi everyone,

I’m currently building an AI agent for our social welfare office to help automate social assistance applications.

image
image348×933 65.2 KB

I’ve run into an issue that suddenly appeared after the agent successfully downloads the first document. The scenario throws the following error:

Error: Failed to process file "tool-file" from external tool response. Error: 400 | Your request cannot be processed due to a validation error. | querystring.fileId Unknown format 'uuid'

Origin: Make

image
image421×813 43.8 KB

I haven’t encountered this specific validation error before. Has anyone seen this or have an idea what might be causing the UUID format issue in the query string? Any help or pointers on how to debug this would be greatly appreciated.

Thanks in advance!

Best regards, Julius

3

Hey Julius,

what is this agent trying to do?

Cause the error sounds like its trying to process a file but instead of the extension its getting the ID.

4

The AI agent is designed to scan a specific folder for documents and extract the corresponding file IDs. Using these retrieved IDs, the agent should then proceed to download the files individually for further processing. The process is expected to handle these IDs correctly and complete the download sequence without validation issues.

The agent successfully initiates the process and downloads the first file from the folder. However, immediately after this first successful step, the process fails for the subsequent files. The system returns a ‘400 Validation Error’ with the message ‘querystring.fileId Unknown format ‘uuid’’, preventing the agent from completing the download for the remaining files.

5

That process sounds very deterministic, why do you need an agent here at all? Just iterate the documents in the folder, get the IDs and go download the files.

But yeah, unknown format most likely refers to file format, and its feeding it the UUID instead. Check which tool its trying to use and what its feeding to it. You can also open a chat with the agent and have it go through the flow step by step to see at which point exactly it fails.

6

Thanks for the input. The reason I’m using an agent here is that it needs to analyze the document content for further steps in the workflow, beyond just the download.

I will try to chat with the agent to identify the error. One thing that has me curious, though: the error origin is listed as ‘Make’ rather than the agent or the OpenAI module itself—is that normal behavior?

7

I’ve tried everything on my end, but I’m still stuck. After debugging the flow, I’ve concluded that the issue must lie either within the AI agent’s internal logic or (more probably) directly within Make’s processing layer, as the agent itself is unable to identify the cause.

I’ve also verified that the file IDs are being retrieved correctly from the ‘Search for files/folders’ module. The strange part is that the first document downloads perfectly every time, but the error only occurs for the subsequent files.

Do you have any other ideas on how to fix this or any unconventional workarounds I could try?

8

I restored an old version of my workflow and now it works just fine again. Thank you for your help anyway!