Flux MCP Server

AI-Powered GitOps

The Flux MCP Server connects AI assistants directly to your Kubernetes clusters, enabling seamless interaction through natural language for troubleshooting, analysis, and operations.

Get Started Documentation

Capabilities

Instant Understanding

Quickly understand your Flux installation status, resource configurations, and deployment histories across environments.

Environment Comparison

Compare Flux configurations for applications and infrastructure between development, staging, and production.

Faster Incident Response

Reduce mean time to resolution (MTTR) during incidents with contextual analysis and actionable remediation steps.

Root Cause Analysis

Automatically correlate events, logs, and configuration changes to identify the source of failures in a GitOps pipeline.

Pipeline Management

Trigger reconciliations, suspend/resume Flux resources, and manage your delivery pipelines with simple requests.

Visualization

Generate diagrams that map out Flux dependencies, resource relationships, and delivery workflows across clusters.

How It Works

AI Agent Claude, Copilot, Gemini
MCP Client
Flux MCP Local or remote
kubeconfig
Kubernetes Your clusters

The Flux MCP Server integrates with AI assistants through the Model Context Protocol, providing them with purpose-built tools to interact with your clusters. When you ask a question or make a request, the AI uses these tools to gather information, analyze configurations, and perform operations based on your instructions.

Security Built-in

The Flux MCP Server is designed with security in mind, ensuring safe interaction between AI assistants and your Kubernetes clusters.

  • Provides a read-only mode for observation without affecting cluster state
  • Masks sensitive information in Kubernetes Secret values
  • Operates with your existing kubeconfig permissions
  • Supports Kubernetes impersonation for limited access
  • Access to the local file system is read-only and restricted to kubeconfig
Public Good Service

Flux Schema MCP Server

The Flux Operator team maintains a free public MCP server that gives AI agents authoritative knowledge of the Kubernetes API. Instead of guessing YAML fields from training data, your assistant queries up-to-date resource definitions and validation rules for any manifest it generates or reviews — no cluster connection needed.

  • Schemas for core Kubernetes, OpenShift, Flux, and 100+ CNCF projects
  • Agents write and review manifests against real schemas, not training data
  • Rebuilt daily to track upstream releases and new API versions
Learn More
json 
{
  "mcpServers": {
    "flux-schema-catalog": {
      "type": "http",
      "url": "https://schemas.fluxoperator.dev/mcp"
    }
  }
}
{
  "mcpServers": {
    "flux-schema-catalog": {
      "type": "http",
      "url": "https://schemas.fluxoperator.dev/mcp"
    }
  }
}

Works with Claude Code, Cursor, VS Code, Codex, and any other MCP client over streamable HTTP.

Official Flux Skills

GitOps AI Skills

The Flux maintainers publish reusable Agent Skills that give coding assistants expertise in Flux CD, Kubernetes, and GitOps best practices. The agent automatically selects the right skill based on the task at hand.

  • gitops-knowledge — answers questions about Flux CD and generates up-to-date YAML for all Flux custom resources
  • gitops-repo-audit — audits GitOps repositories for structure, security, and operational best practices
  • gitops-cluster-debug — troubleshoots Flux on live Kubernetes clusters using the Flux MCP Server
Learn More
shell 
flux plugin install operator

flux operator skills install \
  ghcr.io/fluxcd/agent-skills
flux plugin install operator

flux operator skills install \
  ghcr.io/fluxcd/agent-skills

Install with the Flux CLI in your GitOps repository root, via npx skills, or from the Claude Code plugin marketplace.

Ready for Agentic GitOps?

Install the Flux Operator and connect your AI Agents to Flux MCP.

Get Started Prompting Guide