mcp-server-templates
A flexible platform that provides Docker & Kubernetes backends, a lightweight CLI (mcpt), and client utilities for seamless MCP integration. Spin up servers from templates, route requests through a single endpoint with load balancing, and support both deployed (HTTP) and local (stdio) transports — all with sensible defaults and YAML-based configs.
Documentation
🚀 This Project Has Moved!
> ## ⚠️ **IMPORTANT: This repository has been renamed and moved to MCP Platform**
>
> What changed:
> - New Repository: `Data-Everything/MCP-Platform`
> - New Package: `pip install mcp-platform` (replaces `mcp-templates`)
> - New CLI: `mcpp` command (replaces `mcpt`)
> - Enhanced Features: Improved architecture and expanded capabilities
>
> Migration is easy:
> ```bash
> # Uninstall old package
> pip uninstall mcp-templates
>
> # Install new package
> pip install mcp-platform
>
> # Use new command (all your configs work the same!)
> mcpp deploy demo # instead of mcpt deploy demo
> ```
>
> **📚 Complete Migration Guide | 🆕 New Documentation**
MCP Server Templates (Legacy)
> **⚠️ This version is in maintenance mode. Please migrate to MCP Platform for latest features and updates.**
> Deploy Model Context Protocol (MCP) servers in seconds, not hours.
Zero-configuration deployment of production-ready MCP servers with Docker containers, comprehensive CLI tools, and intelligent caching. Focus on AI integration, not infrastructure setup.
🚀 Quick Start
# Install MCP Templates
pip install mcp-templates
# List available templates
mcpt list
# Deploy instantly
mcpt deploy demo
# View deployment
mcpt logs demoThat's it! Your MCP server is running at `http://localhost:8080`
⚡ Why MCP Templates?
| Traditional MCP Setup | With MCP Templates |
|---|---|
| ❌ Complex configuration | ✅ One-command deployment |
| ❌ Docker expertise required | ✅ Zero configuration needed |
| ❌ Manual tool discovery | ✅ Automatic detection |
| ❌ Environment setup headaches | ✅ Pre-built containers |
Perfect for: AI developers, data scientists, DevOps teams building with MCP.
🌟 Key Features
🖱️ One-Click Deployment
Deploy MCP servers instantly with pre-built templates—no Docker knowledge required.
🔍 Smart Tool Discovery
Automatically finds and showcases every tool your server offers.
🧠 Intelligent Caching
6-hour template caching with automatic invalidation for lightning-fast operations.
💻 Powerful CLI
Comprehensive command-line interface for deployment, management, and tool execution.
🛠️ Flexible Configuration
Configure via JSON, YAML, environment variables, CLI options, or override parameters.
📦 Growing Template Library
Ready-to-use templates for common use cases: filesystem, databases, APIs, and more.
📚 Installation
PyPI (Recommended)
pip install mcp-templatesDocker
docker run --privileged -it dataeverything/mcp-server-templates:latest deploy demoFrom Source
git clone https://github.com/DataEverything/mcp-server-templates.git
cd mcp-server-templates
pip install -r requirements.txt🎯 Common Use Cases
Deploy with Custom Configuration
# Basic deployment
mcpt deploy filesystem --config allowed_dirs="/path/to/data"
# Advanced overrides
mcpt deploy demo --override metadata__version=2.0 --transport httpManage Deployments
# List all deployments
mcpt list --deployed
# Stop a deployment
mcpt stop demo
# View logs
mcpt logs demo --followTemplate Development
# Create new template
mcpt create my-template
# Test locally
mcpt deploy my-template --backend mock🏗️ Architecture
┌─────────────┐ ┌───────────────────┐ ┌─────────────────────┐
│ CLI Tool │───▶│ DeploymentManager │───▶│ Backend (Docker) │
│ (mcpt) │ │ │ │ │
└─────────────┘ └───────────────────┘ └─────────────────────┘
│ │ │
▼ ▼ ▼
┌─────────────┐ ┌───────────────────┐ ┌─────────────────────┐
│ Template │ │ CacheManager │ │ Container Instance │
│ Discovery │ │ (6hr TTL) │ │ │
└─────────────┘ └───────────────────┘ └─────────────────────┘Configuration Flow: Template Defaults → Config File → CLI Options → Environment Variables
📦 Available Templates
| Template | Description | Transport | Use Case |
|---|---|---|---|
| demo | Hello world MCP server | HTTP, stdio | Testing & learning |
| filesystem | Secure file operations | stdio | File management |
| gitlab | GitLab API integration | stdio | CI/CD workflows |
| github | GitHub API integration | stdio | Development workflows |
| zendesk | Customer support tools | HTTP, stdio | Support automation |
🛠️ Configuration Examples
Basic Configuration
mcpt deploy filesystem --config allowed_dirs="/home/user/data"Advanced Configuration
mcpt deploy gitlab \
--config gitlab_token="$GITLAB_TOKEN" \
--config read_only_mode=true \
--override metadata__version=1.2.0 \
--transport stdioConfiguration File
{
"allowed_dirs": "/home/user/projects",
"log_level": "DEBUG",
"security": {
"read_only": false,
"max_file_size": "100MB"
}
}mcpt deploy filesystem --config-file myconfig.json🔧 Template Development
Creating Templates
1. Use the generator:
mcpt create my-template2. Define template.json:
{
"name": "My Template",
"description": "Custom MCP server",
"docker_image": "my-org/my-mcp-server",
"transport": {
"default": "stdio",
"supported": ["stdio", "http"]
},
"config_schema": {
"type": "object",
"properties": {
"api_key": {
"type": "string",
"env_mapping": "API_KEY",
"sensitive": true
}
}
}
}3. Test and deploy:
mcpt deploy my-template --backend mockFull template development guide →
� Migration to MCP Platform
This repository has evolved into MCP Platform with enhanced features and better architecture.
Why We Moved
1. Better Naming: "MCP Platform" better reflects the comprehensive nature of the project
2. Enhanced Architecture: Improved codebase structure and performance
3. Expanded Features: More deployment options, better tooling, enhanced templates
4. Future Growth: Better positioned for upcoming MCP ecosystem developments
What Stays the Same
- ✅ All your existing configurations work unchanged
- ✅ Same Docker images and templates
- ✅ Same deployment workflows
- ✅ Full backward compatibility during transition
Migration Steps
1. Install new package:
pip uninstall mcp-templates
pip install mcp-platform2. Update commands:
# Old command
mcpt deploy demo
# New command (everything else identical)
mcpp deploy demo3. Update documentation bookmarks:
Support Timeline
- Current (Legacy) Package: Security updates only through 2025
- New Platform: Active development, new features, full support
- Migration Support: Available through Discord and GitHub issues
**🚀 Start your migration now →**
�📖 Documentation (Legacy)
- **Getting Started** - Installation and first deployment
- **CLI Reference** - Complete command documentation
- **Template Guide** - Creating and configuring templates
- **User Guide** - Advanced usage and best practices
🤝 Community
- **Discord Server** - Get help and discuss features
- **GitHub Issues** - Report bugs and request features
- **Discussions** - Share templates and use cases
📝 License
This project is licensed under the Elastic License 2.0.
🙏 Acknowledgments
Built with ❤️ for the MCP community. Thanks to all contributors and template creators!
Frequently asked questions
What is mcp-server-templates?
mcp-server-templates is A flexible platform that provides Docker & Kubernetes backends, a lightweight CLI (mcpt), and client utilities for seamless MCP integration. Spin up servers from templates, route requests through a single endpoint with load balancing, and support both deployed (HTTP) and local (stdio) transports — all with sensible defaults and YAML-based configs.
How do I install mcp-server-templates?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is mcp-server-templates open source?
Yes — it is hosted on GitHub at https://github.com/Data-Everything/mcp-server-templates and has 7 stars.
Related MCP tools
Expose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth! Python-based implementation. Trusted by 11000+ developers.
An official Qdrant Model Context Protocol (MCP) server implementation Python-based implementation. Trusted by 1000+ developers.
A middleware to provide an openAI compatible endpoint that can call MCP tools Python-based implementation. Trusted by 800+ developers.
Shell and coding agent on claude desktop app for the Model Context Protocol. Enhance AI assistants with powerful integrations. Python-based implementation.
An MCP server that allows MCP clients like Claude Desktop or Cursor to perform actions in the Unity Editor C#-based implementation.
基于大模型搭建的聊天机器人,同时支持 微信公众号、企业微信应用、飞书、钉钉 等接入,可选择ChatGPT/Claude/DeepSeek/文心一言/讯飞星火/通义千问/ Gemini/GLM-4/Kimi/LinkAI,能处理文本、语音和图片,访问操作系统和互联网,支持基于自有知识库进行定制企业智能客服。
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP