elevy99927/devops-mcp-webui

TL;DR
**Chat with your Kubernetes cluster using natural language!**
This project connects OpenWebUI to Kubernetes, letting you manage your cluster through conversational AI. Ask questions like "What pods are running?" or "Scale my deployment to 5 replicas" and get instant results.
---
Table of Contents
- TL;DR
- Quick Start
- Configure Open-WebUI
- Step 1: Access Open-WebUI
- Step 2: Add the MCP Bridge Tool Server
- Step 3: Add Connections to Models
- Step 4: Verify Tool Integration
- Step 5: Test the Integration
- Troubleshooting Configuration
- Architecture Overview
- Architecture Diagram
- Data Flow
- Network Architecture
- Repository Structure
- Components
- 1. Open WebUI
- 2. Ollama
- 3. MCP-Bridge
- 4. Kubernetes Cluster
- Usage
---
Quick Start
**One-command setup:**
**Componentes installed:**
- Kind (Kubernetes in Docker)
- Open Web UI
- Ollama
- MCP bridge
**After successful installation, you'll see:**
✅ Docker Compose startup complete!
Services available at:
- Open WebUI: http://localhost:3000
- MCP Bridge: http://localhost:9000
- Ollama API: http://localhost:11434
Configure Open WebUI to use the MCP tools
Configure Open-WebUI
After running the installation script, you need to configure Open-WebUI to use the Kubernetes tools:
Step 1: Access Open-WebUI
1. Open your browser and go to **http://localhost:3000**
2. **Create an admin account** on first visit
3. **Sign in** with your new account
Step 2: Add the MCP Bridge Tool Server
1. Click on your **profile icon** (top right)
2. Go to **Settings**
3. Navigate to **Admin Panel** → **Tools**
4. Click **"+ Add Tool Server"**
5. Enter the following details:
- **Name**: Kubernetes Tools
- **URL**: http://mcpo:9000
- **Description**: Kubernetes management via kubectl and helm
6. Click **"Add"**
---
---
Step 3: Add Connections to Models
Configure your AI model connections - use the local Ollama or add remote connections:
1. In **Settings**, go to **Admin Panel** → **Connections**
2. **For Local Ollama** (recommended):
- The local Ollama server should be automatically detected
- Verify connection to http://ollama:11434
3. **For Remote Connections** (optional):
- Add OpenAI, Anthropic, or other model providers
- Configure API keys and endpoints as needed
Step 4: Verify Tool Integration
1. Go back to the **chat interface**
2. Look for the **tools icon** (🔧) in the chat input area
3. You should see available tools like:
- kubectl_get - List Kubernetes resources
- kubectl_apply - Apply manifests
- kubectl_describe - Describe resources
- helm_install - Install Helm charts
- And more...
Step 5: Test the Integration
Try asking these questions to verify everything works (copy and paste into the chat):
**Copy & Paste these queries:**
Troubleshooting Configuration
**If tools don't appear:**
- Check that the MCP Bridge is running: curl http://localhost:9000/health
- Verify the tool server URL is exactly: http://mcpo:9000
- Restart Open-WebUI: docker-compose restart open-webui
**If you get connection errors:**
- Ensure all containers are running: docker-compose ps
- Check container logs: docker-compose logs mcpo
- Verify network connectivity between containers
**Then open http://localhost:3000 and start chatting with your cluster!**
**Example queries:**
- "What pods are running in kube-system?"