Frequently Asked Questions about UBOS MCP Server

Q: What is the Model Context Protocol (MCP)? A: MCP is an open protocol that standardizes how applications provide context to Large Language Models (LLMs), enabling them to interact with external data sources and tools.

Q: What is the UBOS Asset Marketplace’s MCP Server? A: It’s a Flask-based implementation of the Model Context Protocol, designed to enhance LLM capabilities by allowing them to access and use external tools directly.

Q: How does the MCP Server differ from function calling? A: MCP places tool definitions directly in the context window, using natural language, whereas function calling uses structured JSON and API parameters. MCP offers more flexibility.

Q: What are some use cases for the MCP Server? A: Use cases include enhancing AI assistants, improving customer service chatbots, enabling data analysis, robotics automation, and content creation.

Q: What programming language is the MCP Server written in? A: The MCP Server is primarily written in Python, utilizing the Flask framework.

Q: What are the key dependencies for running the MCP Server? A: The key dependencies include Flask, and other libraries specified in the requirements.txt file. An LLM API key and other tool-specific API keys might also be required.

Q: Can I add my own custom tools to the MCP Server? A: Yes, you can add custom tools by creating a new class inheriting from the Tool class, defining parameters, and registering it with the MCP handler.

Q: How does the MCP Server handle conversation flow? A: The server maintains context across multiple interactions by storing and passing conversation history with tool execution results to future LLM requests.

Q: Is the MCP Server suitable for production environments? A: While a development server is provided, production deployment is recommended using a WSGI server like Gunicorn.

Q: Where can I find more information about the UBOS platform? A: Visit the UBOS website at https://ubos.tech to learn more about our full-stack AI Agent development platform.