argoproj-labs/mcp-for-argocd

What it solves

This project provides a way for AI assistants to interact with Argo CD applications using natural language. It bridges the gap between LLM-powered tools (like VS Code, Cursor, or Claude Desktop) and the Argo CD API, allowing users to manage their Kubernetes deployments via a chat interface.

How it works

It implements the Model Context Protocol (MCP) server, which acts as a standardized interface. The server connects to an Argo CD instance using a base URL and an API token. It exposes a set of tools that the AI assistant can call to perform actions in Argo CD, supporting both stdio and HTTP stream transport protocols for flexible integration.

Who it’s for

DevOps engineers and platform engineers who use Argo CD for GitOps and want to integrate their cluster management tasks into their AI-assisted development environment.

Highlights

• Comprehensive Toolset: Includes tools for cluster, project, and application management, as well as resource tree inspection and workload log retrieval.
• Multi-Instance Support: Features a token registry to securely manage credentials for multiple Argo CD instances.
• Flexible Deployment: Supports stdio for local use and HTTP/SSE for network-based deployments, including a stateless mode for Kubernetes HPA.
• Security Controls: Offers a read-only mode to prevent modifications and detailed network exposure settings to control inbound access.