What is MCP?

MCP stands for Model Context Protocol. It’s an open protocol standardizing how applications provide context to Large Language Models (LLMs), enabling them to access external data and tools securely.

What is the MCP Python SDK?

The MCP Python SDK is a toolkit for building MCP-compliant applications. It allows developers to create MCP clients and servers, facilitating seamless communication between LLMs and data sources.

How do I install the MCP Python SDK?

You can install the SDK using pip or uv with the command pip install "mcp[cli]" or uv add "mcp[cli]".

What are Resources in MCP?

Resources are how you expose data to LLMs, similar to GET endpoints in a REST API. They provide data but shouldn’t perform significant computation or have side effects.

What are Tools in MCP?

Tools let LLMs take actions through your server. Unlike resources, tools are expected to perform computation and have side effects, similar to POST endpoints.

What are Prompts in MCP?

Prompts are reusable templates that help LLMs interact with your server effectively. They define interaction patterns and provide structure for LLM requests.

What is the UBOS platform?

UBOS is a Full-stack AI Agent Development Platform that helps you orchestrate AI Agents, connect them with your enterprise data, build custom AI Agents with your LLM model, and create Multi-Agent Systems.

How does the MCP Python SDK integrate with UBOS?

The MCP Python SDK provides the building blocks for context-aware AI, while UBOS offers a platform for orchestrating and managing AI agents. The SDK allows you to connect your AI agents to external data sources and tools, while UBOS provides a visual interface for designing and managing complex AI agent workflows.

Can I use the MCP Python SDK without UBOS?

Yes, the MCP Python SDK can be used independently. It provides the necessary tools for building MCP-compliant applications, regardless of the platform you use for managing your AI agents.

Where can I find the MCP documentation?

You can find the Model Context Protocol documentation at https://modelcontextprotocol.io and the specification at https://spec.modelcontextprotocol.io.