AI Chat
The AI Chat is the core feature of KubeAgentiX, allowing you to interact with your Kubernetes cluster using natural language.
How It Works
- You type a question or command in plain English
- The AI analyzes your request and queries your cluster
- You receive a response with relevant information, commands, or actions
Capabilities
Querying Resources
Ask about any Kubernetes resource:
- Pods, Deployments, StatefulSets, DaemonSets
- Services, Ingresses, ConfigMaps, Secrets
- PersistentVolumes, StorageClasses
- Nodes, Namespaces, and more
Root Cause Analysis
When something goes wrong, ask the AI to investigate:
Why is my app not responding?
The AI will:
- Check pod status and events
- Review recent deployments
- Analyze resource constraints
- Examine logs for errors
Executing Commands
The AI can execute kubectl commands on your behalf:
Delete all evicted pods
You'll always be asked to confirm before any changes are made.
AI Models
KubeAgentiX supports multiple AI providers:
| Provider | Models | Notes |
|---|---|---|
| Google Vertex AI | Gemini 2.5 Flash | Default for Plus/Enterprise |
| OpenAI | GPT-4, GPT-4o | BYOK supported |
| Anthropic | Claude 3.5 | BYOK supported |
| Local | Ollama models | Fully offline |
Privacy
- Your cluster data never leaves your machine when using local models
- With cloud AI providers, only the query context is sent (not your entire cluster state)
- No data is stored or used for training