Introducing Document Robots

Introducing Document Robots

Until now, Maxun helped teams automate the web.

Today, Maxun expands beyond websites and into documents.

We’re excited to introduce Document Robots — a new way to extract structured data from PDFs or convert entire documents into clean, usable formats.

With Document Robots, you can work with PDFs in two ways:

• Extract — Extract specific fields from documents using natural language
• Parse — Convert entire PDFs into clean Markdown, HTML, or extracted links

---

Extract

Upload a sample PDF, describe what information you want, and Maxun builds a reusable extraction robot.

Examples:

• "Extract invoice number, vendor name, and total amount"
• "Extract employee name, joining date, and compensation"
• "Extract client name, contract duration, and payment terms"

When to Use Extract

Use Extract when:

• You need specific fields from invoices, contracts, offer letters, or reports
• You receive similar documents repeatedly
• You want to automate manual document processing

---

Parse

Need the entire document in a clean format?

Upload a PDF and Maxun can instantly convert it into:

• Markdown — Ideal for AI workflows and RAG pipelines
• HTML — Structured and web-friendly
• Links — Every URL found in the document, deduplicated

No prompts. No setup. Just upload and run.

When to Use Parse

Use Parse when:

• You want to feed documents into AI systems
• You need clean text from PDFs
• You want structured document content for downstream automation

---

How It Works

1. Upload a PDF
2. Choose Extract or doc-parse
3. Configure your extraction prompt or output formats
4. Run anytime on new documents

Once created, robots can be reused across unlimited similar files.

---

Using Document Robots with Node SDK

// Extract structured fields from a PDF
const robot = await scraper.createDocument(
  'Invoice Extractor',
  './invoice.pdf',
  {
    type: 'doc-extract',
    prompt: 'Extract invoice number, vendor name, and total amount'
  }
);

const result = await robot.run();
console.log(result.data);

// Parse a PDF into multiple formats
const robot = await scraper.createDocument(
  'Report Parser',
  './report.pdf',
  {
    type: 'doc-parse',
    formats: ['markdown', 'html', 'links']
  }
);

const result = await robot.run();
console.log(result.data.markdown);

Works Everywhere

Document Robots are available through:

• Dashboard (No-Code)
• API
• SDK
• CLI

Try Document Robots

• Documentation: https://docs.maxun.dev/robot/document

Get Started

• https://app.maxun.dev
• https://docs.maxun.dev

Turn PDFs into structured data — now with Maxun.