argocd
MCP server for ArgoCD - GitOps continuous delivery for Kubernetes with GenAI
claude mcp add --transport stdio lokimcpuniverse-argocd-mcp-server python -m argocd_mcp.server \ --env ARGOCD_SERVER="argocd.example.com" \ --env ARGOCD_TIMEOUT="30" \ --env ARGOCD_GRPC_WEB="true" \ --env ARGOCD_INSECURE="false" \ --env ARGOCD_DEV_TOKEN="dev_token" \ --env ARGOCD_AUTH_TOKEN="your_auth_token" \ --env ARGOCD_CLIENT_KEY="/path/to/key.pem" \ --env ARGOCD_DEV_SERVER="argocd-dev.example.com" \ --env ARGOCD_PROD_TOKEN="prod_token" \ --env ARGOCD_CLIENT_CERT="/path/to/cert.pem" \ --env ARGOCD_PROD_SERVER="argocd-prod.example.com"
How to use
This MCP server exposes ArgoCD management capabilities via a Model Context Protocol interface, enabling GenAI-powered tooling to interact with ArgoCD for GitOps-driven Kubernetes deployments. It consolidates ArgoCD operations such as listing and managing applications, projects, repositories, clusters, and application sets, along with status checks, rollbacks, health checks, and RBAC-related tasks. The server authenticates using JWT/API tokens or OIDC/OAuth2 and can be configured for multi-cluster and secret-management workflows. You can invoke a variety of built-in tools exposed by the MCP server to perform common ArgoCD operations from your Claude Desktop or other MCP clients, enabling automated deployment strategies, approvals, and reconciliation tasks across environments.
How to install
Prerequisites:
- Python 3.8+ installed on your system
- Git available to clone sources (optional if using PyPI)
- Access to a running ArgoCD instance (as configured in ARGOCD_SERVER)
Option A: Install from PyPI
pip install argocd-mcp-server
Option B: Install from source
git clone https://github.com/LokiMCPUniverse/argocd-mcp-server.git
cd argocd-mcp-server
pip install -e .
Run the MCP server (as configured in your environment or with a local .env):
export ARGOCD_SERVER=argocd.example.com
export ARGOCD_AUTH_TOKEN=your_auth_token
python -m argocd_mcp.server
Optional: Run via Docker (if you prefer containerized execution):
# Example Docker run (adjust image and environment variables accordingly)
docker run -it --env ARGOCD_SERVER=argocd.example.com --env ARGOCD_AUTH_TOKEN=your_auth_token argocd-mcp-server-image
Additional notes
Environment variables mirror those in the README's Configuration section. Ensure ARGOCD_SERVER and ARGOCD_AUTH_TOKEN are set to valid values for your ArgoCD instance. If you enable ARGOCD_INSECURE, be aware of the security implications in production. The MCP server supports multi-cluster setups via ARGOCD_PROD_SERVER/ARGOCD_DEV_SERVER and their tokens. When using Claude or other MCP clients, provide the appropriate env configuration in your mcpServers entry so the client can authenticate and communicate with ArgoCD correctly. If you encounter connection issues, verify network access to ARGOCD_SERVER, validate tokens, and check ArgoCD's TLS/cert configuration. The available tools cover common ArgoCD operations like listing, syncing, rolling back, and fetching statuses, enabling automated workflows and safe, auditable deployments.
Related MCP Servers
kubectl
A Model Context Protocol (MCP) server for Kubernetes. Install: npx kubectl-mcp-server or pip install kubectl-mcp-server
mcp-for-argocd
An implementation of Model Context Protocol (MCP) server for Argo CD.
k8s
K8s-mcp-server is a Model Context Protocol (MCP) server that enables AI assistants like Claude to securely execute Kubernetes commands. It provides a bridge between language models and essential Kubernetes CLI tools including kubectl, helm, istioctl, and argocd, allowing AI systems to assist with cluster management, troubleshooting, and deployments
jenkins
MCP server for Jenkins CI/CD - Intelligent automation server integration for GenAI
rancher
MCP server for Rancher - Kubernetes management and deployment platform
harness
MCP server for Harness - Continuous delivery and cloud cost management