MCP DeepSeek Update Server

A Model Context Protocol (MCP) server that monitors DeepSeek updates across multiple platforms including official news, GitHub repositories, HuggingFace models, and Twitter/X announcements.

Features

🔍 Multi-Platform Monitoring

• Official DeepSeek news and announcements
• GitHub repository updates and releases
• HuggingFace model releases
• Twitter/X official account updates (@deepseek_ai)

📊 Rich Data Integration

• Real-time RSS feeds for Twitter updates
• GitHub API integration for repository monitoring
• HuggingFace API for model releases
• Smart caching to prevent API rate limits

⚡ MCP Protocol Compliance

• Standard MCP tools for querying updates
• Resource endpoints for accessing cached data
• Configurable time ranges and filtering options
• Error handling and fallback mechanisms

Installation

Option 1: NPX (Recommended - No Installation Required)

# Run directly with npx
npx langgpt

# Initialize configuration
npx langgpt --init

# Test all data sources
npx langgpt --test

# Get Claude Desktop configuration
npx langgpt --config-claude

Option 2: Global Installation

npm install -g langgpt
langgpt --help

Option 3: Build from Source

git clone <repository-url>
cd mcp-deepseek-update-server
npm install
npm run build

Usage

As MCP Server (stdio)

Add to your MCP client configuration:

{
  "mcpServers": {
    "deepseek-updates": {
      "command": "npx",
      "args": ["langgpt"]
    }
  }
}

For Claude Desktop specifically:

# Get the exact configuration
npx langgpt --config-claude

As HTTP Server

# HTTP server not yet implemented
# Use stdio mode for now
npx langgpt

Available Tools

check_deepseek_updates

Get recent DeepSeek updates from all or specific sources.

Parameters:

• days (number, default: 7): Number of days to look back
• source (string, default: "all"): Source to check ("all", "news", "github", "huggingface", "twitter")
• includeMetrics (boolean, default: true): Include engagement metrics

Example:

{
  "name": "check_deepseek_updates",
  "arguments": {
    "days": 14,
    "source": "twitter",
    "includeMetrics": true
  }
}

get_latest_models

Get the latest DeepSeek models from HuggingFace.

Parameters:

• limit (number, default: 10): Number of models to return
• includeStats (boolean, default: true): Include download/like statistics

check_deepseek_news

Check for official DeepSeek news on a specific date.

Parameters:

• date (string): Date in YYMMDD format (e.g., "250601" for 2025-06-01)

get_twitter_updates

Get recent tweets from DeepSeek's official Twitter account.

Parameters:

• count (number, default: 10, max: 50): Number of tweets to fetch
• includeReplies (boolean, default: false): Include reply tweets

Resources

deepseek://updates/{source}

Access cached update data for specific sources.

deepseek://models/latest

Access latest model information from HuggingFace.

deepseek://news/{date}

Access news data for specific dates.

Configuration

Environment variables:

# Optional GitHub token for higher API limits
GITHUB_TOKEN=your_github_personal_access_token

# Cache duration in minutes (default: 15)
CACHE_DURATION=15

# Primary RSS URL for Twitter (default: rsshub.app)
TWITTER_RSS_PRIMARY=https://rsshub.app/twitter/user/deepseek_ai

# Backup RSS URLs (comma-separated)
TWITTER_RSS_BACKUP=https://rsshub.rssforever.com/twitter/user/deepseek_ai

# HTTP server port (default: 3000)
PORT=3000

Data Sources

1. Official News: https://api-docs.deepseek.com/zh-cn/news/
2. GitHub: https://github.com/deepseek-ai organization
3. HuggingFace: https://huggingface.co/deepseek-ai organization
4. Twitter/X: @deepseek_ai via RSShub RSS feeds

Error Handling

The server implements robust error handling:

• Automatic fallback between multiple RSS sources for Twitter
• Rate limit awareness for GitHub API
• Graceful degradation when services are unavailable
• Comprehensive logging for debugging

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

MIT License - see LICENSE file for details.

Support

• Report issues on GitHub
• Check the MCP documentation for integration help
• Join the community discussions