cycloid-mcp-server
Enables AI assistants to interact with the Cycloid platform for infrastructure management, including blueprint discovery, stack creation, pipeline listing, and event monitoring via natural language.
README
Cycloid MCP Server
A Model Context Protocol (MCP) server that provides seamless integration with the Cycloid platform, enabling AI assistants to interact with Cycloid's infrastructure management capabilities through natural language.
Overview
The Cycloid MCP Server bridges the gap between AI assistants and Cycloid's powerful infrastructure automation platform. It enables users to:
- Discover and explore available blueprints and service catalogs
- Create and manage infrastructure stacks using Cycloid's blueprints
- Interact naturally with complex infrastructure workflows through AI assistants
- Leverage Cycloid's expertise in infrastructure as code and automation
Available Tools
CYCLOID_BLUEPRINT_LIST: List all available blueprints with their detailsCYCLOID_BLUEPRINT_STACK_CREATE: Create stacks from blueprints with interactive elicitationCYCLOID_STACKFORMS_VALIDATE: Validate StackForms configuration filesCYCLOID_CATALOG_REPO_LIST: List service catalog repositoriesCYCLOID_EVENT_LIST: List organization events with optional filters (begin,end,severity,type)CYCLOID_PIPELINE_LIST: List all pipelines from Cycloid
Available Resources
cycloid://blueprints: Access to blueprint informationcycloid://service-catalogs-repositories: Access to service catalog repositories informationcycloid://events: Access to recent organization events as JSONcycloid://pipelines: Access to pipeline information
Event Filters
- begin/end: Unix timestamps (strings) delimiting the time window
- severity: One or more of
info,warn,err,crit - type: One or more of
Cycloid,AWS,Monitoring,Custom
Architecture
The server uses a dynamic component registration system based on FastMCP's MCPMixin:
- Automatic Discovery: Components are automatically discovered from
src/components/ - File-based Organization: Components are organized by feature (
catalogs/,stacks/) - Standard Patterns: Each component follows the pattern
*_tools.py,*_resources.py,*_handlers.py,*_prompts.py - MCPMixin Integration: Uses FastMCP's built-in
register_all()method for proper tool/resource registration
HTTP Transport
The Cycloid MCP Server runs as a web service using HTTP transport. Organization and API key are provided via HTTP headers (X-CY-ORG and X-CY-API-KEY) for each request, enabling multi-tenant usage.
Usage: python server.py
Quick Start
Prerequisites
- Python 3.12 or higher
- uv package manager (recommended)
- Docker (for production deployment)
- Valid Cycloid API credentials
Development Setup
# Clone and setup
git clone <repository-url>
cd cycloid-mcp-server
make setup
# Run development server
make dev-server
Production Setup
Using Pre-built Docker Images
The project provides pre-built Docker images via Docker Hub:
# Pull the latest image
docker pull cycloid/cycloid-mcp-server:latest
# Run the server
docker run -p 8010:8010 cycloid/cycloid-mcp-server:latest
Building Locally
# Build Docker image
make build
# Run production server
make prod-server
MCP Configuration
For detailed MCP server configuration examples, see mcp-examples.md.
Available Commands
# Development Environment
make setup # Setup development environment with uv
make install # Install dependencies
make help # Show all available commands
make validate-env # Validate local environment matches CI
# Development Server
make dev-server # Run development server using Python virtual environment
# HTTP Server
python server.py # Run HTTP server
# Production
make build # Build Docker image
make prod-server # Run production server using Docker
# Testing and Quality
make test # Run all tests
make type-check # Run pyright type checking
make lint # Run PEP 8 linting with flake8
make format # Format code with black and isort
make quality-check # Run all quality checks (tests + type checking + linting)
make simulate-ci # Validate environment and run quality checks
# Cleanup
make clean # Clean up development artifacts
make clean-docker # Clean up Docker artifacts
Contributing
See CONTRIBUTING.md for detailed development guidelines.
License
This project is licensed under the MIT License - see the LICENSE file for details.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。