https://github.com/passadis/mcp-sdk-demo
An MCP Server use case with advanced document handling and OpenAI Integration
https://github.com/passadis/mcp-sdk-demo
ai azure document-summarization mcp mcp-server openai
Last synced: 10 months ago
JSON representation
An MCP Server use case with advanced document handling and OpenAI Integration
- Host: GitHub
- URL: https://github.com/passadis/mcp-sdk-demo
- Owner: passadis
- License: mit
- Created: 2025-07-08T00:26:04.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-07-08T14:36:28.000Z (10 months ago)
- Last Synced: 2025-07-08T15:43:00.739Z (10 months ago)
- Topics: ai, azure, document-summarization, mcp, mcp-server, openai
- Language: Python
- Homepage: https://www.cloudblogger.eu/2025/07/08/the-best-way-to-learn-mcp-unlock-the-ai-future/
- Size: 85 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Security: SECURITY.md
Awesome Lists containing this project
README
The best way to Learn MCP: Unlock the AI future
# MCP SDK Implementation
This directory contains a Demo wit Web UI using the actual Model Context Protocol (MCP) SDK.
## Architecture Overview
This implementation uses the official MCP SDK to create proper MCP servers and clients.
### Original Implementation (in parent directories)
- **mcp_server_project**: Mock MCP server for agent registration and message relay
- **agent_a_project**: Frontend agent that initiates document verification/summarization requests
- **agent_b_project**: Backend agent that provides summarization services using Azure OpenAI
### New MCP SDK Implementation
#### Servers (`/servers/`)
- **document_verification_server**: MCP server that provides document verification tools
- **summarization_server**: MCP server that provides text summarization tools using Azure OpenAI
#### Clients (`/clients/`)
- **document_client**: MCP client that uses document verification services
- **summarization_client**: MCP client that uses summarization services
#### Shared (`/shared/`)
- **crypto_utils.py**: Shared cryptographic utilities for secure communication
- **config.py**: Shared configuration and environment variables
## Key Differences from Original
1. **Protocol**: Uses actual MCP protocol instead of custom REST API
2. **Communication**: MCP tools and resources instead of encrypted HTTP requests
3. **Architecture**: Proper MCP server/client model instead of peer-to-peer agents
4. **Security**: MCP-native security features alongside custom encryption
5. **Discoverability**: MCP schema for automatic tool discovery
## Setup Instructions
** Follow the Quickstart Instructions to get started!
1. Install MCP SDK: `pip install mcp`
2. Configure Azure OpenAI credentials in environment variables
3. Run servers: Each server can be started independently
4. Connect clients: Clients can discover and use server capabilities
## Security Model
- Maintains encryption for sensitive data transmission
- Leverages MCP's built-in transport security
- Access control through MCP server authentication
- Public key exchange for document verification
## Migration Benefits
- Standard protocol compliance
- Better tool discovery and schema validation
- Improved error handling and logging
- Enhanced scalability and interoperability
- Ecosystem compatibility with MCP-aware applications