MCP Agents Cloud
@hicodermonkey
About MCP Agents Cloud
MCP Agents Cloud是一个提供管理MCP服务器和代理(Agent)的平台。
Overview
What is MCP Agents Cloud?
MCP Agents Cloud is a self-hosted platform for managing MCP servers and agents. It provides user management, agent SDK (Python package and HTTP API), and streaming chat interactions. Built with Python FastAPI and Vue.js, it requires MySQL 8.0 and is intended for developers or teams who need to orchestrate MCP-based agents.
How to use MCP Agents Cloud?
Install MySQL via Docker (docker-compose up mysql in the docker directory). Set up the backend by creating a Python virtual environment, installing dependencies, copying a YAML configuration file, and running uvicorn main:app --reload. Set up the frontend with npm install then npm run serve for development or npm run build for production. Configuration is managed through environment‑specific YAML files (config-dev.yml, config-test.yml, config-prod.yml).
Key features of MCP Agents Cloud
- User management (registration, login, password reset)
- MCP server CRUD operations
- Agent CRUD and management
- Agent SDK with Python package and HTTP API
- Streaming chat interaction with agents
- SDK API key management
Use cases of MCP Agents Cloud
- Centralized management of multiple MCP servers and configurations
- Creating, editing, and deploying AI agents powered by OpenAI models
- Integrating agents into external applications via the provided SDK
- Team‑based collaborative development of MCP agents with user roles
FAQ from MCP Agents Cloud
What is the difference between MCP Agents Cloud and directly using the MCP protocol?
MCP Agents Cloud adds a management layer with a web UI, user authentication, and an agent SDK. It handles server and agent lifecycle, making it easier to coordinate multiple MCP resources without building those features from scratch.
Which AI models does MCP Agents Cloud support?
The platform’s backend uses the openai-agent framework and includes OpenAI API configuration (key and model settings). It therefore supports OpenAI models as the primary inference engine.
Is MCP Agents Cloud free and open source?
Yes, the project is released under the MIT license. You can deploy it yourself on any infrastructure that meets the requirements (MySQL, Python, Node.js).
How do I set up the required database?
MCP Agents Cloud requires MySQL 8.0. A Docker Compose file is provided in the docker directory – run docker-compose up mysql to start a MySQL instance on port 63307. The data directory (/appdata/share/agents-mcp-cloud/mysql/data) must exist and be owned by your user.
Does MCP Agents Cloud come with API documentation?
Yes. After starting the backend server, API documentation is available at http://localhost:8000/docs (Swagger UI). The configuration also has an api.docs_url setting.
Basic information
More Other MCP clients
Dify Connect MCP
difybaseLlama MCP Streamlit
nikunj2003AI assistant built with Streamlit, NVIDIA NIM (LLaMa 3.3:70B) / Ollama, and Model Control Protocol (MCP).
EasyMCP
mshojaei77A beginner-friendly client for the MCP (Model Context Protocol). Connect to SSE, NPX, and UV servers, and integrate with OpenAI for dynamic tool interactions. Perfect for exploring server connections and chat enhancements.
Dify Plugin Agent Mcp_sse
junjiemDify 1.0 Plugin Support MCP Tools Agent strategies
y-cli 🚀
luohy15A Tiny Terminal Chat App for AI Models with MCP Client Support
Comments