MarkLogic MCP Server

A Python-based MCP (Model-Controller-Processor) server for MarkLogic database operations. This server provides CRUD operations and search functionality for MarkLogic documents through a standardized interface using the MarkLogic REST API.

Features

• Document CRUD operations (Create, Read, Update, Delete)
• Document search with string and structured queries
• Collection management
• Database information retrieval
• TOML-based configuration
• Comprehensive logging
• Full REST API support

Prerequisites

• Python 3.7+
• MarkLogic Server (installed and running)
• Access to a MarkLogic database with appropriate permissions
• REST API instance configured on MarkLogic server

Installation

1. Clone the repository:

git clone https://github.com/yourusername/marklogic-mcp.git
cd marklogic-mcp

2. Install dependencies:

pip install -r requirements.txt

3. Configure your server:

cp pyproject.toml.example pyproject.toml

Edit the pyproject.toml file with your MarkLogic connection details:

[tool.marklogic]
host = "localhost"
port = 8000
user = "your-username"
password = "your-password"

[tool.marklogic.logging]
level = "INFO"
format = "%(asctime)s - %(name)s - %(levelname)s - %(message)s"

Usage

1. Start the MCP server:

python marklogic_mcp_server.py

2. The server provides the following operations:

Create Document

create_document(uri: str, content: dict, collections: list = None)

Read Document

read_document(uri: str)

Update Document

update_document(uri: str, content: dict, collections: list = None)

Delete Document

delete_document(uri: str)

Search Documents

search_documents(
    query: str = None,           # String query (e.g., "marklogic and python")
    structured_query: dict = None,  # Structured query in JSON format
    start: int = 1,             # Starting position
    page_length: int = 10       # Number of results per page
)

Get Database Information

get_database_info()

Example Usage

# Create a document
response = create_document(
    uri="/documents/example.json",
    content={"title": "Example", "content": "This is a test"},
    collections=["test-collection"]
)

# Search using string query
response = search_documents(
    query="marklogic and python",
    start=1,
    page_length=10
)

# Search using structured query
response = search_documents(
    structured_query={
        "query": {
            "term-query": {
                "text": ["marklogic"]
            }
        }
    },
    start=1,
    page_length=10
)

Configuration

The server is configured using pyproject.toml. Here are the available configuration options:

[tool.marklogic]
host = "localhost"      # MarkLogic server host
port = 8000            # MarkLogic server port
user = "admin"         # MarkLogic username
password = "admin"     # MarkLogic password

[tool.marklogic.logging]
level = "INFO"         # Logging level (DEBUG, INFO, WARNING, ERROR, CRITICAL)
format = "%(asctime)s - %(name)s - %(levelname)s - %(message)s"  # Log format

Response Format

All operations return a dictionary with a status field indicating success or failure:

# Success response
{
    "status": "success",
    "message": "Operation result message",
    "content": "Document content (for read operations)",
    "total": "Total results (for search operations)",
    "results": "Search results array",
    "facets": "Search facets (if available)"
}

# Error response
{
    "status": "error",
    "message": "Error message"
}

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.