K8s MCP Server (Go)

io.github.kanegPublished by GitHub account kanegv0.4.6Updated Oct 24, 2026

Safe Kubernetes access for AI agents via MCP. Read-only by default, with explicit permission modes.

Overview

AI-generated overview

A local Kubernetes MCP server that lets an AI assistant inspect clusters read-only by default, with optional readwrite or dangerous permission modes.

What you can do
  • Inspect pods, deployments, StatefulSets, services, ConfigMaps, Secrets, PVCs, Ingress, Jobs, nodes and namespaces, with details and logs
  • Read events, cluster overview, and live node and pod CPU/memory usage
  • Resolve a workload name to matching resources and suggested next tools
  • In readwrite mode: scale, restart, patch or update images on deployments and StatefulSets, create namespaces, check rollout status
  • In dangerous mode: delete pods, deployments, StatefulSets, DaemonSets, namespaces or other resources, and apply arbitrary YAML
When to use it
Useful when you want an assistant to help diagnose Kubernetes problems — unhealthy pods, memory hogs, warning events, failed rollouts — without handing it unrestricted kubectl. Choose readwrite only for controlled operational actions, and dangerous only when full cluster control is truly intended.
Requirements
Runs as a local process on the user's machine (desktop only), installed from a platform-specific MCPB bundle or a native binary; no Docker or Go needed. Needs access to a Kubernetes cluster via a kubeconfig file, the KUBECONFIG environment variable, or in-cluster ServiceAccount credentials. The permission mode is chosen at startup.
Before you install
The dangerous mode allows deleting resources and namespaces and applying arbitrary YAML, which can cause outages; readwrite mode changes workloads. Read-only tools expose ConfigMaps and Secret metadata and keys, so cluster data reaches the model. The server uses your kubeconfig or ServiceAccount credentials, so its permissions are those of that identity.

Installation

In SourceWeft

  1. Open K8s MCP Server (Go) in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

k8s-mcp-go

Safe, read-only-by-default Kubernetes access for AI agents.

A Kubernetes MCP server that lets AI assistants inspect your cluster safely — without giving them unrestricted kubectl.

[GitHub release] [License] [Go] [MCP]

What makes it different

  • Read-only by default: useful for diagnosis without giving AI permission to mutate the cluster.
  • Explicit permission modes: choose readonly, readwrite, or dangerous before the assistant starts.
  • No Docker required: install via MCPB bundles or a single native binary.
  • Kubernetes-native visibility: inspect pods, deployments, services, logs, events, nodes, and live CPU/memory usage.
  • Registry-ready distribution: published as io.github.kaneg/k8s-mcp-go for MCP Registry consumers and packaged clients.

Distribution

Why

Most AI agents can use Kubernetes tools. The real problem is how to stop them from changing the wrong thing.

Giving an AI raw kubectl access is risky. Even a capable model can:

  • delete the wrong resource
  • restart the wrong workload
  • apply a dangerous change
  • turn a debugging session into a production incident

k8s-mcp-go gives AI agents a guardrailed interface to Kubernetes instead of unrestricted shell access.

It is built for the real-world question:

How can I let AI help with Kubernetes, without letting it break my cluster?

Permission Modes

You choose the boundary up front:

What you wantMode
"Let AI inspect and diagnose, but change nothing"readonly
"Allow safe operational actions like scale and restart"readwrite
"Give it full cluster power"dangerous

readonly (default)

For diagnosis, inspection, and safe exploration.

AI can do things like:

  • list pods, deployments, services, nodes, and namespaces
  • read logs and events
  • inspect cluster state
  • check resource usage

It cannot modify workloads or delete resources.

readwrite

For controlled operational workflows.

AI can do things like:

  • scale deployments
  • restart deployments
  • restart statefulsets
  • update images
  • patch deployments
  • create namespaces

It still cannot perform the most destructive operations.

dangerous

Full access.

Use this only when you explicitly want AI to be able to:

  • delete resources
  • delete namespaces
  • apply arbitrary YAML

If you are unsure, use readonly.

Quick Start

Option 1: MCPB Install (Recommended)

Download the .mcpb bundle for your platform from Releases. MCPB-compatible clients (Claude Desktop, Cursor, etc.) can install it directly — no Docker, no Go, no manual setup.

OSArchFile
Linuxx86_64k8s-mcp-go_*_linux_amd64.mcpb
LinuxARM64k8s-mcp-go_*_linux_arm64.mcpb
macOSIntelk8s-mcp-go_*_darwin_amd64.mcpb
macOSApple Siliconk8s-mcp-go_*_darwin_arm64.mcpb
Windowsx86_64k8s-mcp-go_*_windows_amd64.mcpb
WindowsARM64k8s-mcp-go_*_windows_arm64.mcpb

Option 2: Manual Binary Install

Grab the binary archive from Releases and extract it:

bash
# Example: Linux x86_64tar xzf k8s-mcp-go_*_linux_amd64.tar.gzchmod +x k8s-mcp-gosudo mv k8s-mcp-go /usr/local/bin/

Then add it to your MCP client.

Claude Desktop (claude_desktop_config.json):

json
{  "mcpServers": {    "k8s": {      "command": "k8s-mcp-go",      "args": ["-mode=readonly"]    }  }}

Cursor (.cursor/mcp.json):

json
{  "mcpServers": {    "k8s": {      "command": "k8s-mcp-go",      "args": ["-mode=readonly"]    }  }}

Restart your client and start asking questions about your cluster.

Example prompts

Once installed, ask your AI assistant things like:

  • "Show me unhealthy pods in all namespaces."
  • "Which pods are using the most memory?"
  • "Summarize recent warning events."
  • "Inspect this deployment and explain why it is not ready."
  • "Check rollout status for this deployment."

In readonly mode, the assistant can investigate but cannot change the cluster.

Why not just use kubectl?

Because the problem is not whether AI can talk to Kubernetes. The problem is whether it can do so safely.

kubectl is powerful, but it does not give you a product-level permission mode for AI behavior. With k8s-mcp-go, you decide whether the assistant can:

  • inspect only
  • perform limited operational actions
  • or get full control

The permission boundary is the product.

Available Tools (35 total)

Tools are grouped by permission level. Names start with an action verb such as get, list, or delete; top_nodes and top_pods retain Kubernetes' established kubectl top terminology.

Readonly (24)

ToolDescription
get_server_infoShow server version, mode, Kubernetes config source, and runtime details
resolve_workloadResolve an app/workload name to matching resources and suggested next tools
list_podsList pods; supports all_namespaces=true
get_podGet pod details
get_pod_logsGet pod logs
list_deploymentsList deployments; supports all_namespaces=true
get_deploymentGet deployment details
list_statefulsetsList StatefulSets; supports all_namespaces=true
get_statefulsetGet StatefulSet details
list_servicesList services; supports all_namespaces=true
get_serviceGet service details
list_configmapsList ConfigMaps; supports all_namespaces=true
get_configmapGet ConfigMap data
list_secretsList Secrets (keys only); supports all_namespaces=true
get_secretGet Secret metadata and keys
list_pvcList PersistentVolumeClaims; supports all_namespaces=true
list_ingressList Ingress resources; supports all_namespaces=true
list_jobsList Jobs; supports all_namespaces=true
top_nodesCurrent node CPU and memory usage
top_podsCurrent pod CPU and memory usage
list_nodesList cluster nodes
list_namespacesList namespaces
get_cluster_overviewCluster health summary
list_eventsList events

Readwrite (7)

ToolDescription
scale_deploymentScale deployment replicas
restart_deploymentRolling restart a deployment
restart_statefulsetRolling restart a statefulset
set_imageUpdate container image
get_rollout_statusCheck rollout progress
create_namespaceCreate a new namespace
patch_deploymentApply strategic merge patch

Dangerous (7)

ToolDescription
delete_podDelete a pod
delete_deploymentDelete a deployment
delete_statefulsetDelete a StatefulSet and its pods
delete_daemonsetDelete a DaemonSet and its pods
delete_resourceDelete a resource without a dedicated delete tool
delete_namespaceDelete a namespace and all resources
apply_yamlApply arbitrary YAML manifest

Prefer the dedicated delete tools above when one exists. Use delete_resource as the fallback for other kinds. It resolves namespaced versus cluster-scoped resources through Kubernetes discovery.

json
{"api_version":"v1","kind":"ServiceAccount","namespace":"apps","name":"builder"}{"api_version":"rbac.authorization.k8s.io/v1","kind":"ClusterRole","name":"auditor"}

Kubernetes Configuration

Configuration is selected in this order:

  1. The file specified by KUBECONFIG, when set.
  2. The Pod's in-cluster ServiceAccount configuration.
  3. ~/.kube/config when running outside a cluster.

An explicit KUBECONFIG is authoritative. If it cannot be loaded, the server returns an error instead of silently switching to the Pod's ServiceAccount. In-cluster configuration uses Kubernetes' mounted token file so projected ServiceAccount token rotation continues to work without copying tokens into a kubeconfig.

Environment Variables

VariableDescription
KUBECONFIGExplicit kubeconfig path. When unset, use in-cluster configuration or fall back to ~/.kube/config.

Build from Source

bash
git clone https://github.com/kaneg/k8s-mcp-go.gitcd k8s-mcp-gogo build -o k8s-mcp-go .

License

MIT

Source: README.md at commit 889ed38

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.4.6LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts