Back to Home
Azure icon

aks-mcp

Verified Safe

by Azure

Overview

Enables AI assistants to interact with Azure Kubernetes Service (AKS) clusters by translating natural language requests into AKS operations.

Installation

Run Command
docker run -i --rm ghcr.io/azure/aks-mcp:latest --transport stdio

Environment Variables

  • AZURE_TENANT_ID
  • AZURE_CLIENT_ID
  • AZURE_CLIENT_SECRET
  • AZURE_FEDERATED_TOKEN_FILE
  • AZURE_SUBSCRIPTION_ID
  • AZURE_MANAGED_IDENTITY
  • AKS_MCP_COLLECT_TELEMETRY
  • APPLICATIONINSIGHTS_INSTRUMENTATION_KEY
  • USE_LEGACY_TOOLS
  • DISABLE_CACHE

Security Notes

The project demonstrates a strong focus on security, implementing three-tier access control (readonly, readwrite, admin), robust command injection protection using `shlex.Split` and explicit command validation (e.g., restricting binaries to `az`), and secure Azure authentication methods (Service Principal, Managed Identity, Federated Token) with strict path validation for sensitive files. It enforces read-only root filesystems and least-privilege RBAC in Kubernetes deployments via Helm charts. OAuth 2.1 integration with Azure AD includes PKCE, JWT validation, and configurable CORS/redirects. Telemetry includes an opt-out mechanism (`AKS_MCP_COLLECT_TELEMETRY=false`) and device ID hashing. A default Application Insights instrumentation key is present but is not a sensitive access credential and is overridable by an environment variable. While potential tainted input for subprocess launching is acknowledged (`#nosec G204`), the comprehensive validation logic in `azcli/executor.go` and `security/validator.go` is designed to mitigate this risk.

Similar Servers

Stats

Interest Score54
Security Score9
Cost ClassMedium
Avg Tokens1500
Stars101
Forks25
Last Update2025-12-05

Tags

AzureKubernetesAI AssistantModel Context ProtocolCloud Management