Skip to main content
POST
/
v1
/
files
/
presigned-urls
Upload Files
const options = {
  method: 'POST',
  headers: {
    Authorization: '<authorization>',
    'x-organization-id': '<x-organization-id>',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    name: '<string>',
    extension: '<string>',
    mime_type: '<string>',
    size_kb: 123,
    folder_id: '<string>',
    file_ids: ['<string>'],
    status: '<string>',
    project_id: '<string>'
  })
};

fetch('https://api.dcycle.io/v1/files/presigned-urls', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));
{
  "id": "<string>",
  "status": "<string>",
  "url": "<string>",
  "presigned_url": "<string>"
}

​
Upload Files

Use this 3-step handshake to upload files from browsers, scripts, CI jobs, or the Dcycle CLI without streaming file bytes through the API server.
1

Create presigned upload URLs

Call POST /v1/files/presigned-urls. The backend creates pending file rows and returns one presigned S3 URL per file.
2

Upload file bytes to S3

PUT the raw file bytes to each presigned URL using the same Content-Type you sent in step 1.
3

Confirm the upload

Call PATCH /v1/files/batch-update with status=uploaded. This marks the file as uploaded and emits CLASSIFY_DOCUMENT.

​
Step 1: Create Presigned URLs

​
Headers

​
Authorization
string
required
Bearer token or API token for the authenticated user.
​
x-organization-id
string
required
Organization UUID that owns the uploaded files.
​
x-partner
string
Partner header. Use dcycle unless you were given a different integration value.

​
Body

Send an array so one request can prepare multiple uploads.
​
name
string
required
File name without the extension.
​
extension
string
required
File extension such as pdf, csv, xlsx, jpg, or png.
​
mime_type
string
required
MIME type for the uploaded file.
​
size_kb
integer
required
File size in kilobytes.
​
folder_id
string
Optional folder UUID. Omit for root-level uploads.

​
Example

curl -X POST "https://api.dcycle.io/v1/files/presigned-urls" \
  -H "Authorization: Bearer ${DCYCLE_API_KEY}" \
  -H "x-organization-id: ${DCYCLE_ORG_ID}" \
  -H "x-partner: dcycle" \
  -H "Content-Type: application/json" \
  -d '[
    {
      "name": "invoice_january",
      "extension": "pdf",
      "mime_type": "application/pdf",
      "size_kb": 2048
    }
  ]'

​
Response Fields

​
id
string
File UUID stored in PostgreSQL.
​
status
string
Starts as pending.
​
url
string
Final S3-backed file URL stored on the File record.
​
presigned_url
string
Temporary S3 PUT URL used for the direct upload.

​
Step 2: Upload to S3

Upload the raw file bytes to the presigned_url returned in step 1. 
curl -X PUT "${PRESIGNED_URL}" \
  -H "Content-Type: application/pdf" \
  --data-binary @invoice_january.pdf

​
Step 3: Confirm the Upload

PATCH /v1/files/batch-update is the step that flips the file from pending to uploaded and triggers document classification.

​
Body

​
file_ids
string[]
required
File IDs returned by step 1.
​
status
string
required
Use uploaded after a successful S3 upload, or error if the upload failed.
​
project_id
string
Optional project UUID. When present, the backend also creates file_project links.

​
Example

curl -X PATCH "https://api.dcycle.io/v1/files/batch-update" \
  -H "Authorization: Bearer ${DCYCLE_API_KEY}" \
  -H "x-organization-id: ${DCYCLE_ORG_ID}" \
  -H "x-partner: dcycle" \
  -H "Content-Type: application/json" \
  -d '{
    "file_ids": ["11111111-1111-1111-1111-111111111111"],
    "status": "uploaded"
  }'

​
Event Trigger

When status=uploaded, the backend emits CLASSIFY_DOCUMENT with: 
{
  "resource": "POST",
  "method": "/classify-document",
  "file_id": "file-uuid",
  "user_id": "user-uuid",
  "organization_id": "organization-uuid"
}

​
Legacy Alternative

If you do not need the direct-to-S3 flow, POST /v1/files/upload still accepts a standard multipart upload through the backend.