MCP server for intelligent document processing — extract text, classify, and summarize PDFs, images, and documents using vision LLMs.
Extract text from a PDF or image file using OCR.
path(string, required): Path to the document file
Classify a document by type (invoice, report, contract, etc.).
path(string, required): Path to the document file
Generate a structured summary from a document.
path(string, required): Path to the document file
pip install -r requirements.txt
export MCP_BILLING_API=https://mcp-billing-api.onrender.com
uvicorn server:starlette_app --host 0.0.0.0 --port 8000{
"mcpServers": {
"document-intelligence-server": {
"url": "https://mcp-doc-intel.onrender.com/"
}
}
}https://mcp-doc-intel.onrender.com/ — Streamable HTTP transport at root path. Health check at /health.
| Variable | Required | Description |
|---|---|---|
MCP_BILLING_API |
Yes | Billing API endpoint |
MCP_LICENSE_KEY |
Yes | License key for billing |
AGENTICMARKET_SECRET |
No | Secret for AgenticMarket authentication |
MIT