MCP in 10 Minutes

What is MCP in 10 Minutes?

A tutorial and reference implementation that teaches how to build a Model Context Protocol (MCP) server for Yahoo Finance (or any other data source) in roughly 10 minutes. It targets developers who want to learn MCP by creating a practical server.

How to use MCP in 10 Minutes?

Clone the repository, create and activate a virtual environment with uv venv, install dependencies with uv sync, then run the inspector with uv run mcp dev server.py, run the agent with uv run agent.py, and optionally install and run Langflow with uv pip install langflow and uv run langflow run. Ensure Ollama is running.

Key features of MCP in 10 Minutes

• Builds a complete MCP server for Yahoo Finance quickly.
• Uses uv for fast Python package and environment management.
• Includes an MCP inspector for testing and debugging.
• Ships a ready-to-run agent script.
• Integrates with the Langflow visual workflow builder.
• Requires only a local Ollama instance to function.

Use cases of MCP in 10 Minutes

• Learn the Model Context Protocol by building a real server.
• Create a personal finance data tool for stock lookups.
• Prototype and test new MCP server ideas rapidly.
• Serve as a starter template for custom MCP integrations.

FAQ from MCP in 10 Minutes

What dependencies are required?

Ollama must be running locally. The project uses uv for package management; install it separately. Python packages are installed via uv sync.

How do I run the server?

Use uv run mcp dev server.py to start the MCP inspector with the server loaded.

Is there a pre-built agent?

Yes, agent.py is provided and can be run with uv run agent.py after starting Ollama.

Can I use this with Langflow?

Yes, Langflow can be installed and launched after setting up the server. See the Langflow documentation for details.

What license is this under?

This project is licensed under the MIT License.