io.github.sandeepbazar/ocm-mcp-server MCP Server
io.github.sandeepbazar/ocm-mcp-server
Guardrailed multi-cluster Kubernetes operations for AI agents via OCM with policy, approval, and audit.
What is the io.github.sandeepbazar/ocm-mcp-server MCP server?
The ocm-mcp-server MCP server enables AI agents to safely manage multi-cluster Kubernetes environments through Open Cluster Management (OCM). It provides policy-driven operations, approval workflows, and audit logging to ensure guardrailed fleet operations across distributed clusters.
This server bridges AI agents and Kubernetes fleet management by exposing OCM capabilities with built-in safety controls. Use it to orchestrate multi-cluster deployments, enforce policies, require approvals for sensitive operations, and maintain audit trails—all through natural language commands to Claude or other AI agents.
How to install io.github.sandeepbazar/ocm-mcp-server
Copy-paste configuration for popular MCP clients.
KUBECONFIGrequiredPath to the kubeconfig with the OCM hub context (and optional spoke contexts).
OCM_MCP_HUB_CONTEXTKubeconfig context name of the OCM hub cluster (defaults to the current context).
OCM_MCP_READ_ONLYSet to 'true' to disable the gated write path entirely (read-only fleet access).
Use cases
- Deploy applications across multiple Kubernetes clusters with policy enforcement
- Request and approve cluster operations through guardrailed workflows
- Audit and track all fleet management actions performed by AI agents
- Manage cluster policies and compliance across a distributed infrastructure
- Automate multi-cluster operations while maintaining security and governance
io.github.sandeepbazar/ocm-mcp-server MCP server FAQ
It's an MCP server that gives AI agents safe, policy-driven access to manage multiple Kubernetes clusters via Open Cluster Management (OCM), with built-in approval workflows and audit logging.
Yes, ocm-mcp-server is open-source and available on PyPI and as a container image.
Install via PyPI with `pip install ocm-mcp-server` or use the container image `ghcr.io/ocm-mcp-server/ocm-mcp-server`.
Yes, it requires access to an OCM-managed Kubernetes environment and appropriate credentials to connect to your clusters.
The server includes policy enforcement, approval workflows for sensitive operations, and comprehensive audit logging to track all actions.
Yes, as an MCP server, it integrates with any MCP-compatible client including Claude and Cursor.
README (reference)
Source of truth, from the repository.
ocm-mcp-server has moved
This project now lives at github.com/ocm-mcp-server/ocm-mcp-server.
| Code | https://github.com/ocm-mcp-server/ocm-mcp-server |
| Docs | https://ocm-mcp-server.github.io/ |
| PyPI | pip install ocm-mcp-server |
| Container | ghcr.io/ocm-mcp-server/ocm-mcp-server |
| MCP Registry | io.github.ocm-mcp-server/ocm-mcp-server |
Related MCP servers
Search, browse & read 340+ Claude Code marketing & SEO skills from any MCP client.
View repository →
Project Context
Safely manage repository-scoped issues across Linear, GitHub Issues, and Jira Cloud.

Godot MCP server - Godot 4.x engine control via TCP bridge, STL/GLB/OBJ import, GPU particles, an...
GrandOrgue pipe organ simulator MCP server - MIDI bridge, organ control, sample set management, m...
