Document Intelligence MCP - OCR, Summary & Classification
Pricing
Pay per event + usage
Document Intelligence MCP - OCR, Summary & Classification
Real MCP server for Claude Desktop. Extract text (OCR), summarize, classify, and extract tables from document images. Powered by Qwen2.5-VL (Apache 2.0). Connect directly from Claude Desktop via Standby URL.
Pricing
Pay per event + usage
Rating
0.0
(0)
Developer
daehwan kim
Actor stats
0
Bookmarked
1
Total users
0
Monthly active users
7 days ago
Last modified
Categories
Share
Document Intelligence MCP Server
Extract text (OCR), summarize, classify, and extract tables from document images using local AI vision models.
Features
| Tool | Description | Price |
|---|---|---|
extract_text | OCR - Extract all text from document images in markdown format | $0.05/use |
summarize_document | Analyze and summarize document content with key information | $0.10/use |
classify_document | Classify document type (invoice, contract, receipt, etc.) with confidence | $0.05/use |
extract_tables | Extract all tables as structured JSON (headers + rows) | $0.10/use |
Connect via Claude Desktop
Add to your Claude Desktop MCP settings:
{"mcpServers": {"document-intelligence": {"url": "https://ntriqpro--document-intelligence-mcp.apify.actor/mcp?token=YOUR_APIFY_TOKEN"}}}
Supported Document Types
- Invoices and receipts
- Contracts and agreements
- Business letters and memos
- Reports and presentations
- Forms and applications
- ID documents and certificates
- Tables, spreadsheets, and charts
Supported Languages
Responses can be generated in: English, Korean, Japanese, Chinese, Spanish.
Input
Each tool accepts:
imageUrl(required): URL of the document imagelanguage(optional): Response language code (default: "en")
Output
Each tool returns structured JSON with the analysis results.
Example: extract_text
Input:
{"imageUrl": "https://example.com/invoice.png"}
Output:
{"status": "success","text": "# Invoice\n\n**Invoice #**: 2024-001\n**Date**: March 15, 2024\n...","model": "qwen2.5-vl","imageUrl": "https://example.com/invoice.png"}
Example: classify_document
Output:
{"status": "success","classification": {"document_type": "invoice","confidence": 0.95,"language": "en","key_fields": ["invoice_number", "date", "total_amount", "vendor"]}}
Example: extract_tables
Output:
{"status": "success","tables": {"tables": [{"headers": ["Item", "Quantity", "Price"],"rows": [["Widget A", "10", "$5.00"],["Widget B", "5", "$12.00"]]}]}}
Technology
- Vision Model: Qwen2.5-VL (Apache 2.0 License)
- Processing: Local AI inference, zero external API calls
- Privacy: Document images are processed in real-time and not stored
Open Source Licenses
This service uses the following open source models:
- Qwen2.5-VL — Apache 2.0 License
Platform usage is free. You only pay per event (see pricing above).