Ocm MCP Server runs on your own machine — the client starts it, so there is no endpoint to ping. 282 installs a week from pypi. Last commit 29 Aug 2026.
Guardrailed fleet ops for AI agents: multi-cluster Kubernetes via OCM with policy, approval, audit.
Every tool that appeared, vanished or quietly changed what it asks for. Recorded since 29 August 2026. No other catalogue keeps this.
We have not read this server's code yet, 21 h ago · rules 3dff92dd89df
A value the model can set ends up inside a file or shell call. That is not a flaw by itself — for a terminal server it is the job — but it is where things go wrong when it is not.
inj = subprocess.run(["bash", f"{root}/chaos/inject.sh", "failing-rollout", cl],
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
этот файл ставится пользователю, но в репозитории его нет
return subprocess.run(
cmd, shell=True, capture_output=True, text=True, timeout=timeout, check=False
path = os.path.join(REPO, rel_path)
env=dict(os.environ),
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
That is not a flaw by itself — but it is where things go wrong when it is not the job. We re-read this code on every release. Watch it and you hear from us the day another one appears.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add ocm-mcp-server -- uvx ocm-mcp-server
{
"mcpServers": {
"ocm-mcp-server": {
"args": [
"ocm-mcp-server"
],
"command": "uvx"
}
}
}
[mcp_servers.ocm-mcp-server]
command = "uvx"
args = ["ocm-mcp-server"]
{
"mcpServers": {
"ocm-mcp-server": {
"args": [
"ocm-mcp-server"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"ocm-mcp-server": {
"args": [
"ocm-mcp-server"
],
"command": "uvx"
}
}
}
This one needs environment variables set before it will start:
KUBECONFIG (Path to the kubeconfig with the OCM hub context (and optional spoke contexts).), OCM_MCP_HUB_CONTEXT (Kubeconfig context name of the OCM hub cluster (defaults to the current context).), OCM_MCP_READ_ONLY (Set to 'true' to disable the gated write path entirely (read-only fleet access).).
The author declared them in the registry entry; get the values from the project itself.
Guardrailed fleet ops for AI agents: multi-cluster Kubernetes via OCM with policy, approval, audit.
Governed execution cells for AI agents: Docker/K8s sandboxes with policy, audit, and approvals.
Kubernetes MCP server with RBAC-style, context-scoped guardrails for AI agents.
Kubernetes MCP server: 75 tools for GKE, AKS, EKS, RBAC, audit logging, GitOps, multi-cluster.
OpenShift/Kubernetes cluster management via the oc CLI for MCP clients
Kubernetes monitoring & ops for AI agents — safe-by-default access modes and guards.
Safe Kubernetes access for AI agents via MCP. Read-only by default, with explicit permission modes.
MCP server for natural-language Kubernetes operations across single or multi-cluster environments.
Answers built from our own checks of this server.