BREAGEN Logo
Document Understanding

Document Processing API

The Document Processing API provides functionality to manage and analyze documents, including retrieving document details, extracting named entities (NER), answering questions based on document content, generating summaries, converting to markdown, and creating dummy documents.

Document Operations

Document Operations enable comprehensive management and analysis of individual documents within the system.

Document Operations - Get Document GET /v1/api/document/{document_id}

Resource URL/v1/api/document/{document_id}
Response FormatJSON
Requires AuthenticationYes
Rate LimitedYes
HTTPSYes

Description

Retrieve a document's metadata and text by its ID.

Parameters

ParameterDescriptionRequiredTypeDefault Value
document_idPath parameter: ID of the documentYesStringN/A

Example Request

curl -X 'GET' \
'https://braegen.ai/v1/api/document/doc123' \
-H 'accept: application/json'

Example Response

{
  "is_error": false,
  "message": "Document retrieved successfully",
  "results": {
    "document_id": "doc123",
    "meta_data": {},
    "text": "Sample document content..."
  }
}

HTTP Response

HTTP CodeMessage
200Successful Response
422Validation Error

Document Operations - Extract NER POST /v1/api/document/extract_ner

Resource URL/v1/api/document/extract_ner
Response FormatJSON
Requires AuthenticationYes
Rate LimitedYes
HTTPSYes

Description

Extract named entities (NER) from a document.

Parameters (Request Body)

ParameterDescriptionRequiredTypeDefault Value
identifierIdentifier of the document or collectionYesStringN/A
is_documentIndicates if the identifier is a documentNoBooleanfalse

Example Request

curl -X 'POST' \
'https://braegen.ai/v1/api/document/extract_ner' \
-H 'accept: application/json' \
-H 'Content-Type: application/json' \
-d '{
    "identifier": "doc123",
    "is_document": true
}'

Example Response

{
  "is_error": false,
  "message": "NER extraction successful",
  "results": {
    "entities": [
      {"entity": "John Doe", "type": "PERSON"},
      {"entity": "New York", "type": "LOCATION"}
    ]
  }
}

HTTP Response

HTTP CodeMessage
200Successful Response
422Validation Error

Document Operations - Answer Questions On Document POST /v1/api/document/answer

Resource URL/v1/api/document/answer
Response FormatJSON
Requires AuthenticationYes
Rate LimitedYes
HTTPSYes

Description

Answer questions based on the content of specified documents.

Parameters (Request Body)

ParameterDescriptionRequiredTypeDefault Value
userUser identifierYesStringN/A
document_idsList of document IDs to queryYesArrayN/A
questionsList of questions to answerYesArrayN/A

Example Request

curl -X 'POST' \
'https://braegen.ai/v1/api/document/answer' \
-H 'accept: application/json' \
-H 'Content-Type: application/json' \
-d '{
    "user": "user123",
    "document_ids": ["doc123"],
    "questions": ["What is the main topic?"]
}'

Example Response

{
  "is_error": false,
  "message": "Questions answered successfully",
  "results": {
    "answers": [
      {"question": "What is the main topic?", "answer": "The document discusses AI advancements."}
    ]
  }
}

HTTP Response

HTTP CodeMessage
200Successful Response
422Validation Error

Document Operations - Document Summary GET /v1/api/document/{document_id}/summary

Resource URL/v1/api/document/{document_id}/summary
Response FormatJSON
Requires AuthenticationYes
Rate LimitedYes
HTTPSYes

Description

Provide a summary of the specified document.

Parameters

ParameterDescriptionRequiredTypeDefault Value
document_idPath parameter: ID of the documentYesStringN/A

Example Request

curl -X 'GET' \
'https://braegen.ai/v1/api/document/doc123/summary' \
-H 'accept: application/json'

Example Response

{
  "is_error": false,
  "message": "Summary generated successfully",
  "results": {
    "summary": "This document outlines key features of AI technology."
  }
}

HTTP Response

HTTP CodeMessage
200Successful Response
422Validation Error

Document Operations - Document Markdown GET /v1/api/document/{document_id}/markdown

Resource URL/v1/api/document/{document_id}/markdown
Response FormatJSON
Requires AuthenticationYes
Rate LimitedYes
HTTPSYes

Description

Retrieve the document content in markdown format.

Parameters

ParameterDescriptionRequiredTypeDefault Value
document_idPath parameter: ID of the documentYesStringN/A
start_pageQuery parameter: Starting pageNoInteger1
end_pageQuery parameter: Ending pageNoInteger1

Example Request

curl -X 'GET' \
'https://braegen.ai/v1/api/document/doc123/markdown?start_page=1&end_page=2' \
-H 'accept: application/json'

Example Response

{
  "is_error": false,
  "message": "Markdown generated successfully",
  "results": {
    "markdown": "# Document Title\n\nSample content..."
  }
}

HTTP Response

HTTP CodeMessage
200Successful Response
422Validation Error

Document Operations - Create Dummy Document POST /v1/api/document/create-dummy-doc

Resource URL/v1/api/document/create-dummy-doc
Response FormatJSON
Requires AuthenticationYes
Rate LimitedYes
HTTPSYes

Description

Create a dummy document for testing purposes.

Parameters

ParameterDescriptionRequiredTypeDefault Value
document_idQuery parameter: ID of the documentYesStringN/A
collection_idQuery parameter: ID of the collectionYesStringN/A

Example Request

curl -X 'POST' \
'https://braegen.ai/v1/api/document/create-dummy-doc?document_id=doc123&collection_id=col456' \
-H 'accept: application/json'

Example Response

{
  "is_error": false,
  "message": "Dummy document created successfully",
  "results": {
    "document_id": "doc123",
    "collection_id": "col456"
  }
}

HTTP Response

HTTP CodeMessage
200Successful Response
422Validation Error