MCP Client (Claude, IDE, etc.)
        │
        │ MCP Protocol (JSON-RPC over stdio)
        │
        ▼
┌─────────────────────────────┐
│   MCP Document Server       │
├─────────────────────────────┤
│   Tools:                    │
│   ├── extract_text          │
│   ├── chunk_document        │
│   ├── search_chunks         │
│   ├── summarize_document    │
│   └── get_metadata          │
├─────────────────────────────┤
│   Resources:                │
│   ├── document://{path}     │
│   └── chunks://{doc_id}     │
├─────────────────────────────┤
│   Parsers:                  │
│   ├── PDFParser             │
│   ├── DocxParser            │
│   └── TextParser            │
└─────────────────────────────┘

git clone https://github.com/BabyChrist666/mcp-document-server.git
cd mcp-document-server

python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

pip install -r requirements.txt

{
  "mcpServers": {
    "document-analysis": {
      "command": "python",
      "args": ["-m", "mcp_doc_server"],
      "cwd": "/path/to/mcp-document-server"
    }
  }
}

claude --mcp-server "python -m mcp_doc_server"

{
  "name": "extract_text",
  "arguments": {
    "file_path": "/path/to/document.pdf"
  }
}

{
  "name": "chunk_document",
  "arguments": {
    "file_path": "/path/to/document.pdf",
    "chunk_size": 500,
    "overlap": 50
  }
}

{
  "name": "search_chunks",
  "arguments": {
    "query": "What are the payment terms?",
    "doc_id": "contract_2024",
    "top_k": 5
  }
}

{
  "name": "summarize_document",
  "arguments": {
    "file_path": "/path/to/report.pdf",
    "detail_level": "brief"
  }
}

{
  "name": "get_metadata",
  "arguments": {
    "file_path": "/path/to/document.pdf"
  }
}

pytest tests/ -v