Cloud Native MCP Server

🇨🇳 中文文档 | 🇬🇧 English

A high-performance Model Context Protocol (MCP) server for Kubernetes and cloud-native infrastructure management with 11 integrated services and 250+ tools.

---

Features

• 11 Integrated Services: Kubernetes, Grafana, Prometheus, Loki, Kibana, Elasticsearch, Helm, Alertmanager, Jaeger, OpenTelemetry, Utilities
• 250+ MCP Tools: Comprehensive toolset for infrastructure operations
• Multi-Protocol Support: SSE and streamable-http modes
• Smart Caching: LRU cache with TTL support for optimal performance
• Performance Optimized: JSON encoding pool, response size control, intelligent limits
• Enhanced Authentication: API Key (with complexity requirements), Bearer Token (JWT validation), Basic Auth
• Secrets Management: Secure credential storage and rotation
• Input Sanitization: Protection against injection attacks
• Audit Logging: Track all tool calls and operations
• LLM-Optimized: Summary tools and pagination to prevent context overflow

LLM Tool Calling Tips

• Prefer summary and paginated tools first, then switch to full-detail tools only when needed.
• Send structured JSON objects and arrays when a parameter represents an object or list; many handlers still accept legacy JSON strings for compatibility.
• Use RFC3339 timestamps for Prometheus and tracing time fields.
• For Kubernetes, omit namespace for cluster-scoped resources and include it for namespaced resources.
• Some MCP client wrappers already unwrap tool results into objects. Inspect the raw return value before calling JSON.parse(...).
• For Kibana, some handlers accept both camelCase and snake_case parameter aliases, but the schema name is still the preferred form.

---

Services Overview

Service	Tools	Description
kubernetes	34	Core container orchestration and resource management
helm	34	Application package management and deployment
grafana	43	Visualization, monitoring dashboards, and alerting
prometheus	20	Metrics collection, querying, and monitoring
loki	7	LogQL queries, label discovery, and stream inspection
kibana	73	Log analysis, visualization, and data exploration
elasticsearch	12	Log storage, search, and data indexing
alertmanager	16	Alert rules management and notifications
jaeger	8	Distributed tracing and performance analysis
opentelemetry	9	Metrics, traces, and logs collection and analysis
utilities	6	General-purpose utility tools

Total: 250+ tools

---

Quick Start

Binary

# Download the latest release
curl -LO https://github.com/mahmut-Abi/cloud-native-mcp-server/releases/latest/download/cloud-native-mcp-server-linux-amd64
chmod +x cloud-native-mcp-server-linux-amd64

# Run in SSE mode (default)
./cloud-native-mcp-server-linux-amd64 --mode=sse --addr=0.0.0.0:8080

Docker

docker run -d \
  --name cloud-native-mcp-server \
  -p 8080:8080 \
  -v ~/.kube:/root/.kube:ro \
  mahmutabi/cloud-native-mcp-server:latest

From Source

git clone https://github.com/mahmut-Abi/cloud-native-mcp-server.git
cd cloud-native-mcp-server

make build
./cloud-native-mcp-server --mode=sse --addr=0.0.0.0:8080

---

API Endpoints

SSE Mode

Endpoint	Description
/api/aggregate/sse	All services (recommended)
/api/kubernetes/sse	Kubernetes service
/api/helm/sse	Helm service
/api/grafana/sse	Grafana service
/api/prometheus/sse	Prometheus service
/api/loki/sse	Loki service
/api/kibana/sse	Kibana service
/api/elasticsearch/sse	Elasticsearch service
/api/alertmanager/sse	Alertmanager service
/api/jaeger/sse	Jaeger service
/api/opentelemetry/sse	OpenTelemetry service
/api/utilities/sse	Utilities service

Streamable-HTTP Mode

Use --mode=streamable-http to expose MCP streamable HTTP endpoints such as:

• /api/aggregate/streamable-http

SSE Smoke Test

Validate SSE handshake and initialize end-to-end against a running server:

# without auth
make sse-smoke BASE_URL=http://127.0.0.1:8080

# with API key auth
API_KEY=your-key make sse-smoke BASE_URL=http://127.0.0.1:8080

---

Documentation

• Complete Tools Reference - Tool guide aligned with the current server inventory
• Configuration Guide - Configuration options and examples
• Deployment Guide - Deployment strategies and best practices
• Security Guide - Authentication, secrets management, and security best practices
• Architecture Guide - System architecture and design
• Performance Guide - Performance features and tuning

---

Build

# Build for current platform
make build

# Run tests
make test

# Code linting
make lint

# Docker build
make docker-build

---

Contributing

Contributions are welcome! Please read our contributing guidelines and submit pull requests.

---

License

MIT License - see LICENSE for details.