MarkLogic
Connects large language models to live MarkLogic data, enabling natural language querying through a Model Context Protocol interface.
About
This read-only Model Context Protocol (MCP) server allows large language models (LLMs) like Claude Desktop to query live MarkLogic data. It leverages the CData JDBC Driver for MarkLogic, which exposes MarkLogic data as relational SQL models. The server wraps this driver, providing a simple MCP interface that enables LLMs to retrieve live information by asking natural language questions, eliminating the need for direct SQL interaction.
Key Features
- Supports direct execution of SQL SELECT queries via the `run_query` tool
- Presents MarkLogic data as relational SQL models via CData JDBC Drivers
- Facilitates natural language querying of MarkLogic data by LLMs
- Offers a standardized MCP interface for data access
- Provides tools for retrieving database schema (tables, columns)
- 0 GitHub stars
Use Cases
- Connecting LLM-based assistants (e.g., Claude Desktop) to MarkLogic databases
- Developing AI applications that require real-time data access from MarkLogic without writing SQL
- Enabling non-technical users to extract insights from MarkLogic using natural language