📄 Smart File Analyzer & Extractor

Turn any document, image, or text file into structured data or concise summaries instantly.

This Actor serves as a universal interface for file processing. Whether you need to digitize receipts, summarize long reports, or extract specific data points from PDFs, this tool adapts to your needs dynamically.

🚀 Key Features

• Universal File Support: Works with PDFs, Images (JPG, PNG), Text files, and more.
• Two Powerful Modes:
  • Auto-Summary: Get a comprehensive summary, key insights, and categorization automatically.
  • Custom Extraction: Provide a simple schema, and the AI will extract exactly the data fields you need (e.g., "Invoice Number", "Total Amount", "Date").
• Cost-Effective: You only pay for the files you process and the data generated.
• Strict Output: Guarantees valid JSON output, making it easy to integrate into your database or workflows.

---

🛠️ How It Works

1. Summary Mode (Default)

If you provide only a file URL, the Actor will analyze the document and return a standardized summary object containing:

• Title
• Executive Summary
• Key Insights (Bullet points)
• Category & Language

2. Extraction Mode (Advanced)

If you provide an outputSchema, the Actor transforms into a dedicated data entry assistant. It scans the file and extracts only the specific fields you defined, returning clean JSON ready for your API or database.

---

💡 Use Cases

Use Case	Input	Output
Expense Management	Upload a photo of a receipt	JSON with { "vendor": "Starbucks", "total": 12.50, "date": "2024-01-01" }
Legal Review	Upload a 50-page contract	A concise summary with "Key Risks" and "Expiration Dates" highlighted.
HR Automation	Upload a candidate's Resume (PDF)	JSON with { "name": "John Doe", "skills": ["JS", "Python"], "experience_years": 5 }
Content Moderation	Upload user-generated images	JSON with { "is_safe": true, "description": "A cat sitting on a sofa" }

---

📥 Input Configuration

fileUrl (Required)

The direct URL to the file you want to process.

• Max file size: 100 MB.

outputSchema (Optional)

A JSON object defining the data you want to extract. If left empty, the Actor runs in Summary Mode.

Example Schema (for an Invoice):

{
  "type": "object",
  "properties": {
    "invoice_id": { "type": "string", "description": "The unique invoice number" },
    "total_amount": { "type": "number", "description": "Final total including tax" },
    "vendor_name": { "type": "string", "description": "Name of the service provider" },
    "line_items": {
      "type": "array",
      "items": { "type": "string" },
      "description": "List of items purchased"
    }
  },
  "required": ["invoice_id", "total_amount"]
}