XPack -Marketplace
The world’s first open-source MCP monetization platform, to quickly create and sell your own MCP server in just minutes. | XPack 是全球首个开源 MCP 交易平台,帮助你在10分钟内快速搭建自己的 MCP 商店并立刻开始销售 MCP 服务。
claude mcp add --transport stdio xpack-ai-xpack-mcp-marketplace docker run -i xpackai/xpack-mcp-market
How to use
XPack is a full-featured MCP marketplace that lets you deploy and operate an MCP market where developers can publish and monetize their MCP services. The stack includes a backend, frontend, and various services (MySQL, Redis, RabbitMQ) managed behind a single deployment. The marketplace provides one-click OpenAPI to generate MCP service configurations, an SEO-friendly homepage and service pages, built-in billing (per-call or token-based), user accounts, Stripe payments, and sign-in via email or Google OAuth. To use it, deploy the provided Docker image as described, then access the admin interface to manage services, billing, and users. The frontend UI will be served by the container, while the backend APIs handle service provisioning, user management, and transactions.
Once running, you can create, configure, and publish MCP services through the admin UI, test generated OpenAPI configurations, and integrate payments via Stripe for monetization. Customers can discover services via the marketplace, sign in, and subscribe or pay per call depending on the configured billing model. Administrative tasks such as user management, order tracking, and service updates are accessible from the admin panel.
How to install
Prerequisites:
- Docker and Docker Compose installed on your host
- Basic familiarity with pulling and running containerized services
Step-by-step:
-
Ensure runtime dependencies are available (MySQL, Redis, RabbitMQ) as described in the deployment docs. Start these services or use the provided docker-compose setup.
-
Pull and run the market container:
- You can deploy via Docker directly using: docker run -i xpackai/xpack-mcp-market
-
Access the marketplace UI and admin panel:
- Admin UI: http://<host>:8000/admin
- Public marketplace: http://<host>:8000/
-
If you plan to use local frontend/backend builds, follow the Build section in the README to build frontend assets and start the backend services as separate processes (UVicorn for Python backend, etc.).
-
Configure environment variables as needed for production (e.g., database connection, API URL, and payment settings) via your chosen deployment method (Docker environment variables, Kubernetes secrets, etc.).
Additional notes
Tips and common issues:
- Ensure MySQL, Redis, and RabbitMQ are reachable by the marketplace backend before starting the container.
- If you use Docker Compose, ensure the port mappings match your host (the docs map port 8000 to 80 inside the container; adjust as needed).
- For production, set up proper environment variables for NEXT_PUBLIC_API_URL, database credentials, and Stripe keys.
- If the frontend cannot reach the backend, verify the API URL configuration and network connectivity between containers.
- Regularly update the container image to incorporate security patches and new features from the XPack MCP Market project.
Related MCP Servers
cui
A web UI for Claude Code agents
mcp-agent-langchainjs
Serverless AI agent using LangChain.js and Model Context Protocol (MCP) integration to order burgers from a burger restaurant
hackmd
A Model Context Protocol server for integrating HackMD's note-taking platform with AI assistants.
mcp -js
MCP server that exposes YepCode processes as callable tools for AI platforms. Securely connect AI assistants to your YepCode workflows, APIs, and automations.
codemesh
The Self-Improving MCP Server - Agents write code to orchestrate multiple MCP servers with intelligent TypeScript execution and auto-augmentation
mcp-reporter
mcp-reporter is a streamlined utility that generates comprehensive capability reports for Model Context Protocol servers, empowering developers to easily understand available functionality across their MCP servers ecosystem for both documentation and integration into other tools.