MCP Context SDK

Overview

The MCP Context SDK is a modular, production-ready Python SDK that enables AI agents, models, and tools to work with structured context schemas. It supports dynamic context construction, prompt generation, and integration with modern AI frameworks.

Features

• Dynamic context construction and validation
• Schema-to-prompt conversion
• Integration with LangChain, AutoGen, FastAPI, CrewAI, LangGraph, and OpenDevin
• CLI and Streamlit UI demos
• Type hints and schema validations

Quick Start

Installation

pip install mcp-context-sdk

Usage

CLI Tool

List available schemas:

mcp list-schemas coding

Convert context to prompt:

mcp convert-context context.json coding --version v1

Streamlit UI

Run the Streamlit app to interactively edit and preview context:

streamlit run examples/streamlit_ui.py

How to Use with LangChain/AutoGen

• Integrate MCP context into LangChain and AutoGen workflows for enhanced AI operations.

Creating Your Own Schema

• Define domain-specific schemas and integrate them into the MCP SDK.

Running the Landing Page

To view the landing page locally, open docs/landing/index.html in your browser. For deployment, consider using GitHub Pages or Vercel.

License

This project is licensed under the MIT License.

Contribution

Contributions are welcome! Please read the contribution guidelines first.