Help: error message: "message": "Site_Data in input form must be a file",

[FierdCS]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:)
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

I'm sending the request
{
"inputs": {

    "Site_Data": [
        {
            "transfer_method": "local_file",
            "upload_file_id": "hiding id for security",
            "type": "document"
        }
    ]
},
"response_mode": "streaming",
"user": "abc-123"

}

and getting the response:
{
"code": "invalid_param",
"message": "Site_Data in input form must be a file",
"status": 400
}

how do I fix?

2. Additional context or comments

No response

[FierdCS]

I have confirmed a file upload and am getting a id in the response but this id does not help me contact the workflow

[dosubot[bot]]

Hi @FierdCS! I'm Dosu and I’m helping the dify team.

The file upload ID by itself isn't enough to connect your file to the workflow. Dify expects file inputs like Site_Data to be provided as a mapping (dictionary) with fields such as transfer_method, upload_file_id, and type. For workflows, you should include your uploaded file's metadata in the files array of your workflow API request, not just in the inputs field.

Here's how your payload should look:

{
  "inputs": {
    "Site_Data": "your_file_variable_name" // or the variable name expected by your workflow
  },
  "files": [
    {
      "transfer_method": "local_file",
      "upload_file_id": "your-uploaded-file-id",
      "type": "document"
    }
  ],
  "response_mode": "streaming",
  "user": "abc-123"
}

This way, Dify can validate and process the file correctly for your workflow. The key is to reference the file mapping in the files array, not just pass the ID in inputs—the backend uses this mapping to build the file object and link it to your workflow step. For more details, see the relevant code logic here and here.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}    

[FierdCS]

I tried it that way but still got this issue is there anything else I may be missing?

[FierdCS]

[FierdCS]

here is my workflow

[dosubot[bot]]

Thanks for sharing your workflow details. Based on the code and recent updates, here’s what you need to check:

1. Make sure the START node’s Site_Data input is set to type File in the workflow editor. If it’s not, file inputs won’t be accepted and you’ll get the error you’re seeing.
2. When calling /v1/workflows/run, your payload should use "Site_Data" as a single dictionary (not an array or string), like:

   {
     "inputs": {
       "Site_Data": {
         "type": "document",
         "transfer_method": "local_file",
         "upload_file_id": "your-uploaded-file-id"
       }
     },
     "files": [],
     "response_mode": "streaming",
     "user": "abc-123"
   }

   Do not wrap the file input in square brackets.
3. After changing input types or workflow structure, publish updates for each node in the web interface so the workflow recognizes the new configuration.
4. If you’re self-hosting, double-check your file upload and environment config (like FILES_URL) so files are accessible to the workflow.

Recent backend changes now strictly validate the transfer_method field and filter out invalid file mappings, so your input must match the expected format exactly. If you follow these steps and still see the error, double-check that your API request matches the structure above and that the workflow input is set to type File in the editor.
See details in the recent PR and file model validation.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}