# Rancher MCP for AI Agents AI Agent Connect

> Rancher MCP lets you manage Kubernetes clusters and container workloads directly through your AI client. It handles cluster oversight, namespace discovery, and pod introspection across your entire Rancher deployment. Stop toggling between tabs and start using natural language to query your infrastructure.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_48JQqDtai3S1zaUsKSh5UlUL9fX8wRsUCkOFQCNW/ai-agent-connect
- **Tags:** kubernetes, container-orchestration, cluster-management, devops, containers, infrastructure-monitoring

## Description

Managing a sprawling Kubernetes environment usually means jumping between multiple browser tabs, running endless commands, and squinting at complex dashboards to find out why a specific pod is crashing. This Connector changes that by letting your AI agent do the heavy lifting for you. Instead of hunting for IDs or manually checking every node's status, you can just ask your agent to give you a high-level view of your clusters or drill down into a specific project's health. It connects your control plane to your chat interface, giving you a direct line to your container infrastructure. You can check if your microservices are actually running, see what's happening in a specific namespace, or find out which users have access to your platform without ever leaving your workspace. It's a much faster way to handle routine infrastructure checks. You can get a clear picture of your deployment topology or verify that your latest updates reached the correct nodes without having to manually parse through logs. If you're looking for a way to simplify your day-to-day operations, this Connector is one of the highlights in the Vinkius catalog for infrastructure management. You get to stop being a human bridge between your AI and your cloud provider and start actually getting work done, whether you're troubleshooting a production outage or just doing a routine audit of your container environment.

## Tools

### get_cluster
Retrieve the specific details for a single Kubernetes cluster. Get deep insights into a specific environment quickly.

### get_project
Get the full details for a specific Rancher project. This helps you understand the scope of your specific project area.

### list_apps
See all Helm applications installed in a specific project. This is the best way to inventory your deployed software.

### list_catalogs
See all available Helm chart repositories in your environment. This helps you keep track of your available software sources.

### list_clusters
Get a list of every Kubernetes cluster managed by your Rancher instance. Use this to see your full fleet at a glance.

### list_namespaces
Find the namespaces associated with a specific project. This helps you navigate logical partitions without manual config.

### list_nodes
See every node currently active within a specific cluster. This is useful for checking your underlying hardware status.

### list_users
See every user account currently registered in your Rancher platform. Use this to audit your security and access.

### list_workloads
See all deployments and statefulsets within a project. This provides a quick overview of your running services.

### list_projects
Find the project IDs and logical projects within a cluster. This makes it easy to group your resources logically.

## Prompt Examples

**Prompt:** 
```
Show me all the clusters I have running in Rancher right now.
```

**Response:** 
```
I've checked your Rancher environment. Here are your active clusters:

| Cluster Name | Status |
| --- | --- |
| production-us-east | Active |
| staging-development | Provisioning |
```

**Prompt:** 
```
Are there any issues with the pods in the frontend-tier namespace?
```

**Response:** 
```
I've checked the `frontend-tier` namespace. All pods in that namespace are currently running normally. No crash loops or restarts were detected in the recent logs for your workloads.
```

**Prompt:** 
```
Who has user accounts in our Rancher platform?
```

**Response:** 
```
I've pulled the user list for you. You currently have 4 active accounts:

| User | Role |
| --- | --- |
| admin_user | Cluster Admin |
| dev_team_1 | Project Member |
| qa_tester | Read Only |
| support_lead | Project Member |
```

## Capabilities

### List all managed clusters
See every Kubernetes cluster connected to your Rancher control plane in one go.

### Explore namespaces
Find specific logical partitions within your clusters without manually editing config files.

### Monitor workloads
Check the status of deployments and statefulsets across your different environments.

### Inspect individual pods
Get real-time data on pod states and identify crashing containers instantly.

### View Helm applications
See all installed Helm apps within a specific project to manage your software stack.

### Audit user accounts
See who has access to your Rancher platform to keep your environment secure.

## Use Cases

### Debugging a crashing microservice
A backend dev notices an error and asks the agent to find the 'auth-service' pod in the 'backend-production' namespace. The agent uses list_workloads to find it and reports the status instantly.

### Auditing cluster access
A security lead needs to know who has access to the production environment. They ask the agent to list all users in Rancher to verify permissions.

### Checking cluster health
A DevOps engineer wants to see if the new staging cluster is up. They ask the agent to list all clusters and check the status of the 'staging-dev' instance.

### Inventorying Helm apps
A manager wants to see what's running in the 'frontend' project. The agent uses list_apps to provide a full list of installed Helm applications.

## Benefits

- Faster debugging: Use list_workloads to see if your services are up without searching through the UI.
- Reduced context switching: Check pod states with list_nodes and list_namespaces without opening a terminal.
- Better security: Audit access quickly with list_users to ensure only the right people have cluster access.
- Simplified Helm management: View your software stack using list_apps to see what's actually deployed.
- Rapid discovery: Find project IDs and cluster details with get_project and get_cluster in seconds.

## How It Works

The bottom line is you get a text-based interface for your entire Kubernetes infrastructure.

1. Connect the Rancher MCP to your AI client.
2. Provide your Rancher URL and a Bearer Token for API access.
3. Ask your agent to query clusters, check pod health, or list namespaces.

## Frequently Asked Questions

**Can I use Rancher MCP to manage my Kubernetes clusters?**
Yes, this Connector lets you use your AI client to view and interact with all the clusters managed by your Rancher control plane.

**Does the Rancher MCP work with my existing Rancher setup?**
It does. You just need to provide your Rancher URL and a Bearer Token to connect your agent to your current infrastructure.

**Can I use this to check my pod status?**
Yes, you can ask your agent to check specific pods or list all workloads in a project to see if they are running correctly.

**How does Rancher MCP help with DevOps tasks?**
It removes the need to manually click through the Rancher UI. You can query cluster health, list namespaces, and check Helm apps using natural language.

**Can I see which users have access?**
Yes, the Connector can list all user accounts in your Rancher platform, making it easy to audit your security settings quickly.

**Does Rancher MCP support Helm charts?**
It can list the available Helm chart repositories and show you which Helm applications are currently installed in your projects.

**How do I generate a Bearer Token in Rancher?**
Log into your Rancher UI. Go to your User Profile menu and select 'API & Keys'. Create a new API Key with appropriate scopes. Combine the generated Access Key and Secret Key using a colon (`AccessKey:SecretKey`), or use the final Bearer Token provided to authenticate the Connector.

**Does the integration require direct cluster network access?**
No, the Connector acts exclusively by pinging the overarching main Rancher Server API URL rather than connecting locally to standard cluster API endpoints separately. Your network configuration remains secure as it relies completely on the Rancher gateway role.

**Can I deploy new workloads to my cluster via the chat interface?**
Currently, the server allows listing and fetching comprehensive statuses for namespaces, workloads, endpoints, clusters, and pods natively to observe the infrastructure state interactively without allowing unmanaged destructive configurations to roll out purely based on prompt.