Documents

Add PDF document

Uploads a PDF into a knowledge base and queues it for ingestion. Send multipart/form-data with a knowledge_base_id and a file. The file must be a valid PDF (content-checked), max 20 MB. The document is created with status QUEUED and processed asynchronously — text extraction → chunking → embedding → vector indexing; poll GET /api/ext/document/{document_id} for status. Returns the created document. Only PDF is supported today (/url and /text are not yet available).

POST

https://api.nabrah.ai/api/ext/document/file

Add PDF document

curl --request POST \
  --url 'https://api.nabrah.ai/api/ext/document/file' \
  --header 'X-API-Key: <your-api-key>'

{
  "id": "00000000-0000-4000-8000-000000000000",
  "knowledge_base_id": "00000000-0000-4000-8000-000000000000",
  "type": "file",
  "status": "unknown",
  "title": "string",
  "summary": "string",
  "filename": "string",
  "created_at": "2026-01-15T12:00:00.000Z",
  "updated_at": "2026-01-15T12:00:00.000Z"
}

Form fields (multipart/form-data)

knowledge_base_id

The target knowledge base id to attach the document to.

knowledge_base_id

file

The PDF file to ingest. Must be a valid PDF (verified by content), maximum 20 MB.

file

The document returns immediately as QUEUED and is embedded in the background — it becomes retrievable once its status reaches DONE.

Authorization

X-API-Keystringheaderrequired

Your workspace API key. Create one in the dashboard under API keys.

Body

multipart/form-data

Response

201

Successful Response

application/json

422

Validation Error

application/json