Reference

Universal Document Extractor

Overview

Heavstal Universal Extractor is a multi-format parsing engine. It accepts a direct URL to a document or code file and returns the clean, raw text content. This is ideal for feeding documents into LLMs or search indexes.

Supported Formats

PDF (.pdf)

Word (.docx)

Plain Text (.txt)

Code (.js, .ts, .py, .java, .html, .css, .json, .md, etc.)

Endpoint

POST /doc-extract

Request Parameters

Field	Type	Required	Description
url	string	Required	Direct URL to the file.

Example Usage

javascript

      const res = await fetch('https://heavstal-tech.vercel.app/api/v1/doc-extract', {
        method: 'POST',
        headers: { 
          'Content-Type': 'application/json',
          'x-api-key': 'YOUR_API_KEY' 
        },
        body: JSON.stringify({ url: 'https://example.com/contract.docx' })
      });

Successful Response

json

      {
        "status": "success",
        "creator": "HEAVSTAL TECH",
        "data": {
          "detected_type": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
          "extension": "docx",
          "content": "Contract Agreement\nThis agreement is made between..."
        }
      }