mcpcan
MCPCAN is a centralized management platform for MCP services. It deploys each MCP service using a container deployment method. The platform supports container monitoring and MCP service token verification, solving security risks and enabling rapid deployment of MCP services. It uses SSE, STDIO, and STREAMABLEHTTP access protocols to deploy MCP。
claude mcp add --transport stdio kymo-mcp-mcpcan docker run -i mcpcan-image \ --env REGISTRY_PREFIX="Optional: switch between global/CN mirrors (set in .env before running)"
How to use
MCPCan is an open-source platform for managing MCP (Model Context Protocol) services with a modern web interface. It provides unified MCP service lifecycle management, multi-protocol compatibility and conversion, real-time monitoring, built-in security/authentication, and one-stop deployment. The deployment workflow is containerized to avoid host conflicts and to support Kubernetes-based environments. Users interact with MCPCan via its Web UI to monitor services, convert between MCP protocols, and deploy MCP services across environments. The system is designed for DevOps teams to centrally manage multiple MCP service instances, view status and performance, and securely control access to resources.
Once deployed, you can start/stop services, configure MCP pipelines, and leverage protocol conversion to bridge different MCP architectures. The platform supports Kubernetes deployments for scaling and provides a straightforward path to rollout via Helm charts in supported environments. Access the Web UI to manage credentials, permissions, and service configurations, and use the built-in monitoring dashboards to observe service health and throughput across your MCP landscape.
How to install
Prerequisites
- Docker and Docker Compose installed on your host (or a Kubernetes cluster for Helm deployment)
- Git for cloning repositories
Step-by-step installation (Docker Compose quick start)
-
Get deployment repository
- For international users: git clone https://github.com/Kymo-MCP/mcpcan-deploy.git cd mcpcan-deploy/docker-compose/
- For China users (Gitee mirror): git clone https://gitee.com/kymomcp/mcpcan-deploy.git cd mcpcan-deploy/docker-compose/
-
Prepare configuration
- Copy example.env to .env to set environment variables (e.g., REGISTRY_PREFIX to switch mirrors)
-
Generate final configuration (as needed by the deployment) chmod +x replace.sh ./replace.sh
-
Start services (Docker Compose) docker compose up -d
-
Access the MCPCan Web UI Open http://localhost (or http://<your-public-ip>) and login with: Username: admin Password: admin123
Kubernetes/Helm deployment (if using Helm charts)
- Refer to the Helm Deployment Guide: https://kymo-mcp.github.io/mcpcan-deploy/
- Follow the Helm charts installation steps to deploy MCPCan components onto a Kubernetes cluster.
Additional notes
Tips and common issues:
- Ensure Docker is running before starting MCPCan services. If you encounter port conflicts, modify the .env configuration to adjust ports/interfaces used by MCPCan components.
- The REGISTRY_PREFIX environment variable can help switch between global and China mirrors when pulling images; set this in the .env file before running replace.sh.
- Access the Web UI at http://localhost or the assigned IP address; default login is admin/admin123 unless overridden by deployment configuration.
- For Kubernetes deployments, use the Helm charts described in the deployment guide and ensure your cluster has sufficient resources for the MCPCan components (Web UI, Backend, Gateway, Market, and Authz).
- This MCP server supports multi-protocol compatibility and protocol conversion; explore the UI to configure MCP pipelines and observe real-time monitoring dashboards for service health.
Related MCP Servers
ida-pro
AI-powered reverse engineering assistant that bridges IDA Pro with language models through MCP.
mcp-security-hub
A growing collection of MCP servers bringing offensive security tools to AI assistants. Nmap, Ghidra, Nuclei, SQLMap, Hashcat and more.
mcp-graphql
Model Context Protocol server for GraphQL
mcp-for-argocd
An implementation of Model Context Protocol (MCP) server for Argo CD.
git
An MCP (Model Context Protocol) server enabling LLMs and AI agents to interact with Git repositories. Provides tools for comprehensive Git operations including clone, commit, branch, diff, log, status, push, pull, merge, rebase, worktree, tag management, and more, via the MCP standard. STDIO & HTTP.
ContextPods
Model Context Protocol management suite/factory. An MCP that can generate and manage other local MCPs in multiple languages. Uses the official SDKs for code gen.