This directory contains tutorials on the Model Context Protocol (MCP), an open protocol for seamless integration between LLM applications and external data sources.
Each tutorial in this directory includes its own setup and installation instructions. Please refer to the individual tutorial files for specific requirements.
Common requirements:
- Python 3.10 - 3.13
- IBM watsonx.ai account
- Install dependencies (see Installation above)
- Navigate to this directory:
cd tutorials/11-model-context-protocol - Open and run your first tutorial
A custom MCP server implementation for IBM tutorials.
- Topics: MCP server development, custom integrations
- Prerequisites: Node.js, TypeScript
- Estimated time: 45-60 minutes
Note: The MCP Server Integration with IBM Bob tutorial has been moved to the IBM Bob category for better organization alongside other IBM Bob tutorials.
The Model Context Protocol is an open standard that enables:
- Standardized communication between LLMs and external systems
- Context sharing across different tools and data sources
- Tool integration with a unified interface
- Data source connectivity through a common protocol
- Interoperability: Connect any LLM to any data source
- Standardization: Common protocol reduces integration complexity
- Flexibility: Easy to add new tools and data sources
- Security: Controlled access to external resources
- Scalability: Efficient context management
┌─────────────────┐
│ LLM Client │ (Your application)
└────────┬────────┘
│ MCP Protocol
┌────────▼────────┐
│ MCP Server │ (Integration layer)
└────────┬────────┘
┌────┴────┐
┌───▼──┐ ┌──▼───┐
│ Tool │ │ Data │
│ │ │Source│
└──────┘ └──────┘
- Resources: External data sources (files, databases, APIs)
- Tools: Functions that can be invoked
- Prompts: Pre-defined prompt templates
- Context: Shared information between client and server
- Database Integration: Query databases through standardized interface
- File System Access: Read and process files
- API Integration: Call external APIs
- Document Processing: Extract and analyze document content
Found an issue or want to add a new MCP tutorial? See our Contributing Guide.
See the LICENSE file in the repository root.