An open-source AI agent that brings the power of Gemini directly into your terminal.
MCP Operator
Production Kubernetes operator for deploying and validating MCP servers
Install
git clone https://github.com/vitorbari/mcp-operatorInstall command taken from the project's README or registry entry. Always review code before running it.
README excerpt
Read full README on GitHub →MCP Operator
⚠️ Alpha Software
This project is in early development. APIs may change, features may be incomplete, and bugs are expected. We'd appreciate your feedback via issues.
A Kubernetes operator for deploying MCP servers.
Why use this?
Protocol validation - The operator connects to your MCP server after deployment and verifies it actually speaks MCP. It detects which protocol your server uses (SSE or Streamable HTTP), what capabilities it advertises, and whether it requires authentication. This catches configuration mistakes early - wrong port, wrong path, server not actually running MCP, etc.
Correct transport configuration - SSE and Streamable HTTP have different requirements. The operator handles the transport-specific configuration (paths, session management, keep-alive settings) so you don't have to figure out the right Service annotations or health check paths for each protocol type.
Observability - If you have Prometheus Operator installed, the operator creates ServiceMonitors and Grafana dashboards for your MCP servers. There's also an optional metrics sidecar that can collect MCP-specific metrics (request counts, latencies, etc.)
Standard Kubernetes resources - Under the hood, it creates Deployments, Services, ServiceAccounts, and HPAs. Nothing proprietary.
What this doesn't do
- Ingress/external exposure - Creates a ClusterIP Service by default. You need to create your own Ingress, Gateway, or change the Service type to LoadBalancer if you want external access.
- Authentication - The operator detects if your server requires auth, but doesn't handle authentication itself. You need to configure auth at your server or ingress layer.
- stdio transport - Only supports HTTP-based transports (SSE and Streamable HTTP). If your MCP server uses stdio, you can wrap it with an adapter like supergateway that exposes it as SSE or Streamable HTTP, then deploy that container with this operator. See the containerizing guide for details.
- MCP client - This deploys servers, not clients. It doesn't help you connect to MCP servers from your applications.
- TLS termination - Doesn't configure TLS by default. You can enable TLS termination via the metrics sidecar (
spec.sidecar.tls) if you're using metrics, or use an ingress controller.
Quick Start
See the Getting Started Guide for a complete walkthrough.
Installation
Two options:
Option 1: Install via Helm (Recommended)
# Get latest version
VERSION=$(curl -s https://api.github.com/repos/vitorbari/mcp-operator/releases | jq -r '.[0].tag_name' | sed 's/^v//')
# Install
helm install mcp-operator oci://ghcr.io/vitorbari/mcp-operator \
--version ${VERSION} \
--namespace mcp-operator-system \
--create-namespace
Option 2: Install via kubectl
# Get latest version
VERSION=$(curl -s https://api.github.com/repos/vitorbari/mcp-operator/releases | jq -r '.[0].tag_name')
# Install from release assets
kubectl apply -f https://github.com/vitorbari/mcp-operator/releases/download/${VERSION}/install.yaml
Helm is easier to configure. kubectl has fewer dependencies.
Your First MCPServer
Create my-server.yaml:
Similar plugs
The fastest path to AI-powered full stack observability, even for lean teams.
⭐AI-driven public opinion & trend monitor with multi-platform aggregation, RSS, and smart alerts. 告别信息过载,你的 AI 舆情监控助手与热点筛选工具!聚合多平台热点 + RSS 订阅,支持关键词精准筛选。AI 智能筛选新闻 + AI 翻译 + AI…
An open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Enhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini,…