Frequently Asked Questions about MCP Server

Q: What is an MCP Server? A: MCP stands for Model Context Protocol. An MCP Server acts as an intermediary, providing tools for AI models (like LLMs) to connect to and interact with various database systems using a standardized protocol.

Q: Which databases does the MCP Server support? A: The MCP Server supports SQLite, PostgreSQL, MySQL/MariaDB, and SQL Server databases.

Q: How do I install the MCP Server? A: You can install the MCP Server from source using pip install -e . after cloning the repository.

Q: How do I configure the MCP Server? A: You can configure the server using environment variables or a JSON configuration file. The configuration file allows you to specify connection details for different databases.

Q: Can I use the MCP Server with LLMs other than Claude? A: Yes, the MCP Server can be run as a standalone web server, allowing other LLMs (like Llama 3) to connect via HTTP.

Q: What kind of tools does the MCP Server provide? A: The MCP Server provides tools for connection management (add, test, list, remove), query execution (execute, get, insert, update, delete), schema management (list, get, create, drop, alter), and transaction management (begin, commit, rollback).

Q: How can I add a new database connection? A: You can add a new connection by providing the connection details (connection ID, type, and database-specific parameters) in a JSON format.

Q: How can I execute a SQL query? A: You can execute a SQL query by providing the connection ID, the query string, and optional parameters in a JSON format.

Q: How can I manage database schemas using the MCP Server? A: You can use the schema management tools to list tables, get table schemas, create new tables, drop tables, create indexes, and drop indexes.

Q: Does the MCP Server support transactions? A: Yes, the MCP Server supports transaction management, allowing you to begin, commit, and rollback transactions to ensure data integrity.

Q: How do I connect to the MCP Server from another LLM? A: Other LLMs can connect to the MCP Server via HTTP by making POST requests to the /list_tools and /call_tool endpoints.

Q: What are the benefits of using the MCP Server with UBOS? A: The MCP Server simplifies data access for AI Agents, reduces development time, improves data accuracy, enhances AI Agent capabilities, and seamlessly integrates with the UBOS platform for AI Agent development.

Q: Where can I find more information about the MCP Server? A: You can find more information on the UBOS Asset Marketplace: https://ubos.tech.