### Install RubixKube Observer

Source: https://docs.rubixkube.ai/environments/aws

Run this command on the machine where you want to install the Observer. It downloads and executes the installation script.

```bash
curl -fsSL https://api.rubixkube.ai/install/observer.sh | bash -s -- --api-key=rk_YOUR_API_KEY
```

--------------------------------

### Initial Cluster Cost Overview Queries

Source: https://docs.rubixkube.ai/tutorials/chat-cost-analysis

Start by asking for a general cost estimate and infrastructure summary to get a high-level view of your cluster's expenses.

```text
"What's my cluster cost estimate?"
"Show me infrastructure summary"
```

--------------------------------

### Install RubixKube Observer on KIND

Source: https://docs.rubixkube.ai/environments/kubernetes

Create a KIND cluster and then apply the Observer installation manifest. This is useful for local development and evaluation.

```bash
kind create cluster --name rubixkube-test
kubectl apply -f "https://api.rubixkube.ai/install/observer.yaml?apiKey=rk_YOUR_API_KEY"
```

--------------------------------

### Example Troubleshooting Conversation

Source: https://docs.rubixkube.ai/tutorials/chat-troubleshooting

This example demonstrates a multi-turn conversation for diagnosing a pod failure, from initial query to applying a fix.

```text
You: "Why did memory-hog-demo fail?"
Agent: [Provides RCA: OOMKilled due to 50Mi limit]

You: "Why was the limit set to 50Mi?"
Agent: [Checks pod spec: "That's what was configured in the YAML"]

You: "What's a good limit for stress containers?"
Agent: [Suggests 150Mi with 50% buffer]

You: "Show me how to apply that"
Agent: [Provides kubectl command or YAML patch]
```

--------------------------------

### Ask for Comparisons

Source: https://docs.rubixkube.ai/tutorials/chat-advanced

Provides examples of queries to compare resource usage or versions between different services or deployments.

```text
"Compare memory usage: dev vs. prod"
"Which uses more CPU: api-gateway or checkout?"
"Show me the diff between v1.2 and v1.3"
```

--------------------------------

### Install RubixKube Observer

Source: https://docs.rubixkube.ai/environments/kubernetes

Apply the Observer installation manifest to your Kubernetes cluster. Ensure you use the API key from your RubixKube workspace.

```bash
kubectl apply -f "https://api.rubixkube.ai/install/observer.yaml?apiKey=rk_YOUR_API_KEY"
```

--------------------------------

### Summarize Production Health

Source: https://docs.rubixkube.ai/cli/examples

Start your day with a single prompt against production to get a summary of its health. This can be aliased in your shell startup file for convenience.

```bash
rubix chat --prompt "Summarise production health: open incidents, pending actions, deployments in the last 12 hours."
```

```bash
# In ~/.zshrc or ~/.bashrc
alias morning-health='rubix chat --prompt "Summarise production health: open incidents, pending actions, deployments in the last 12 hours."'
```

--------------------------------

### SKILL.md File Format Example

Source: https://docs.rubixkube.ai/using/skills

This is an example of the SKILL.md file format, which includes YAML frontmatter for metadata and a markdown body for instructions. Ensure your skill files adhere to this structure for proper validation.

```markdown
---
name: my-custom-skill
description: A clear description of what this skill does and when to use it.
allowed-tools: tool_one tool_two tool_three
---

## Instructions

Step 1: Do the first thing
  Action: Describe what to do
  Tool: tool_one

Step 2: Do the next thing
  Action: Describe the follow-up action
  Tool: tool_two

```

--------------------------------

### Afternoon Pre-Deployment Workflow

Source: https://docs.rubixkube.ai/tutorials/chat-advanced

A streamlined workflow for validating and initiating deployments. Includes risk assessment and monitoring setup.

```text
"Status of payment-service?"
→ Current state, recent issues

[Upload new-deployment.yaml]
"Validate this"
→ Security, resource, config checks

"What's the risk?"
→ Blast radius analysis

"Deploying now - monitor it"
→ Agent watches for issues
```

--------------------------------

### Install Rubix CLI Globally with Yarn or pnpm

Source: https://docs.rubixkube.ai/cli/installation

Alternative methods for global installation using Yarn or pnpm package managers.

```bash
yarn global add @rubixkube/rubix
# or
pnpm add -g @rubixkube/rubix
```

--------------------------------

### Right-Sizing Guidance Queries

Source: https://docs.rubixkube.ai/tutorials/chat-cost-analysis

Get specific recommendations for optimal CPU and memory limits for common applications like Nginx or Redis, or based on a pod's image.

```text
"What's a good memory limit for nginx?"
"How much CPU does Redis typically need?"
"Recommend optimal limits based on my pod's image"
```

--------------------------------

### Install Rubix CLI Globally with npm

Source: https://docs.rubixkube.ai/cli/installation

Recommended method for installing the Rubix CLI globally. Requires Node.js 18 or later.

```bash
npm install -g @rubixkube/rubix
```

--------------------------------

### Install Observer Agent

Source: https://docs.rubixkube.ai/support/troubleshooting

Apply the Observer agent configuration to register a new cluster. Ensure you replace YOUR_KEY with your actual API key.

```bash
kubectl apply -f "https://api.rubixkube.ai/install/observer.yaml?apiKey=YOUR_KEY"
```

--------------------------------

### Structured Skill Instructions

Source: https://docs.rubixkube.ai/using/skills

Example of how to structure skill instructions using clear, step-by-step formatting. Each step includes an action and the specific tool to be used.

```text
Step 1: Obtain High-Level Cluster Overview
  Action: Begin by getting a broad snapshot of the
  entire cluster's state.
  Tool: fetch_infrastructure_snapshot

Step 2: Analyze Resource Health
  Action: Identify any resources in error, warning,
  or degraded states.
  Tool: analyze_kubernetes_resources
```

--------------------------------

### Describe Newly Created EC2 Instances

Source: https://docs.rubixkube.ai/environments/aws

List EC2 instances tagged as 'observer' that were provisioned by the installer. This helps verify the creation and status of a fresh EC2 instance.

```bash
aws ec2 describe-instances \
  --filters "Name=tag:rubixkube,Values=observer" \
  --query 'Reservations[].Instances[].{ID:InstanceId,Status:State.Name,Name:Tags[?Key==`Name`]|[0].Value}'
```

--------------------------------

### Kubernetes Concept Learning Query

Source: https://docs.rubixkube.ai/tutorials/chat-advanced

Use this query to deepen your understanding of Kubernetes concepts by requesting explanations with real-world examples from your cluster.

```bash
"Explain [concept] with examples from my cluster"
```

--------------------------------

### Install and use Node.js LTS with nvm

Source: https://docs.rubixkube.ai/cli/troubleshooting

If you use nvm (Node Version Manager), this command installs and switches to the latest LTS Node.js version, which is recommended for the Rubix CLI.

```bash
# With nvm
nvm install --lts
nvm use --lts
```

--------------------------------

### Multi-line Query Example

Source: https://docs.rubixkube.ai/tutorials/chat-advanced

Use Shift+Enter to create multi-line queries for complex questions, allowing for more detailed or structured input.

```bash
Shift+Enter
```

--------------------------------

### Typical Questions for ImagePullBackOff

Source: https://docs.rubixkube.ai/tutorials/chat-troubleshooting

Example questions to ask Chat when a pod is in ImagePullBackOff state, helping to diagnose image pulling issues.

```text
"What's wrong with broken-image-demo?"
"Why can't it pull the image?"
"Show me the events for broken-image-demo"
```

--------------------------------

### Ask RubixKube a question in Teams

Source: https://docs.rubixkube.ai/integrations/microsoft-teams

Mention the bot with a question to get information about your infrastructure. Replies will be posted in the same thread.

```text
@RubixKube what is the current health of the payments service?
```

--------------------------------

### Example Agent Response for Pod Logs

Source: https://docs.rubixkube.ai/tutorials/chat-troubleshooting

Illustrates the agent's output after retrieving and analyzing pod logs, highlighting key messages and explaining the root cause of the issue.

```text
Recent logs from memory-hog-demo:

dispatching hogs: 0 cpu, 0 io, 1 vm, 0 hdd
stress: info: [1] dispatching hogs: 0 cpu, 0 io, 1 vm, 0 hdd
stress: dbug: [1] using backoff sleep of 3000us
stress: dbug: [1] setting mem-stress worker 1 forked

This shows the container is intentionally consuming memory ("1 vm"),
which is causing it to exceed the 50Mi limit and get OOMKilled.
```

--------------------------------

### Complete Conversation Summary - Initial Queries

Source: https://docs.rubixkube.ai/tutorials/chat-cost-analysis

This section summarizes the initial queries used in a complete cost analysis conversation, starting with cluster-wide estimates.

```text
1. "What's my cluster cost estimate?"
2. "Show me infrastructure summary and cost reduction"
3. "Which pods are most expensive?"
4. "How can I reduce costs for heavy-app-1 and heavy-app-2?"
```

--------------------------------

### Get Help for Commands

Source: https://docs.rubixkube.ai/cli/commands

The --help flag can be used to print help information for any command. This is useful for understanding command usage and available options.

```bash
rubix --help
rubix chat --help
rubix model --help
```

--------------------------------

### Example Chat Queries

Source: https://docs.rubixkube.ai/tutorials/chat-basics

These are examples of natural language questions you can ask the SRI Agent to get infrastructure intelligence.

```natural-language
"What pods are failing?"
"Why is memory-hog-demo crashing?"
"Show me resource usage"
```

--------------------------------

### Example Questions for Chat Integration

Source: https://docs.rubixkube.ai/using/insights

Examples of questions you can ask the AI Chat based on incident context provided from Insights.

```text
"Explain this incident in simple terms"
"How do I implement recommendation #1?"
"Has this pod failed before?"
"What are similar incidents in the past?"
```

--------------------------------

### Start Rubix CLI Chat

Source: https://docs.rubixkube.ai/cli/commands

Starts an interactive chat session. Can be used to resume an existing session or pre-send a prompt.

```bash
rubix chat [options]
```

--------------------------------

### Monthly Cost Optimization Routine - Week 1 Query

Source: https://docs.rubixkube.ai/tutorials/chat-cost-analysis

Begin the monthly routine by identifying all over-provisioned pods to pinpoint areas for potential cost savings.

```text
"Show me all over-provisioned pods"
```

--------------------------------

### Junior Developer Learning Queries

Source: https://docs.rubixkube.ai/tutorials/chat-advanced

Ask these questions to learn about Kubernetes concepts and your cluster's applications. The agent explains concepts using your cluster's context.

```text
"What is a pod?"
"Explain CrashLoopBackOff in simple terms"
"Why does Kubernetes kill OOM pods?"
"What's the difference between Deployment and Pod?"
"How do resource limits work?"
```

```text
"What applications are deployed here?"
"How does payment-service connect to the database?"
"What technologies are we using?"
"Show me an example of a healthy pod"
```

```text
"If I delete this pod, what happens?"
"What would breaking this service impact?"
"Is it safe to restart api-gateway?"
```

--------------------------------

### Verify Rubix CLI Installation

Source: https://docs.rubixkube.ai/cli/installation

Command to check the installed version of the Rubix CLI. Ensure your global npm bin directory is in your PATH if the command is not found.

```bash
rubix --version
```

--------------------------------

### Start Rubix CLI Chat with Prompt

Source: https://docs.rubixkube.ai/cli/commands

Starts an interactive chat session and immediately sends a predefined prompt. This mode is non-interactive when a session ID is also provided.

```bash
rubix chat --prompt "What is the health of production?"
```

--------------------------------

### Example Monitoring Questions for Chat

Source: https://docs.rubixkube.ai/tutorials/monitor-infrastructure-health

Use these prompts to quickly query monitoring data via the Chat interface. They help in understanding recent changes, identifying high-error resources, checking resource utilization, and tracking deployment issues.

```text
What changed in the payments service in the last hour?
Which resources have the highest error rate today?
Any hosts above 80% memory right now?
Show me deployments that rolled back in the last 24 hours.
```

--------------------------------

### Evening Post-Deploy Check Workflow

Source: https://docs.rubixkube.ai/tutorials/chat-advanced

Verify the stability and performance of a new deployment in the evening. This check takes about 1 minute.

```text
"How's the new payment-service?"
→ Pod status, errors, metrics

"Resource usage vs. old version?"
→ Before/after comparison

"Any regressions?"
→ Error rate, latency check
```

--------------------------------

### Install Node.js LTS with Homebrew

Source: https://docs.rubixkube.ai/cli/troubleshooting

If your Node.js version is older than 18, you may need to upgrade. This command shows how to install the latest LTS version using Homebrew on macOS.

```bash
# macOS with Homebrew
brew install node@20
```

--------------------------------

### Update Rubix CLI Globally

Source: https://docs.rubixkube.ai/cli/installation

Command to update the globally installed Rubix CLI to the latest version.

```bash
npm update -g @rubixkube/rubix
```

--------------------------------

### Ask Your First Question in RubixKube

Source: https://docs.rubixkube.ai/getting-started/quickstart

Once your environment is healthy, use the Chat feature to ask operational questions about your services and resources. Examples include checking service health, identifying restarted resources, or monitoring host resource utilization.

```text
What is the health of the payments service?
Which resources restarted most in the last 24 hours?
Any hosts running above 80% memory right now?
```

--------------------------------

### Launch Rubix CLI

Source: https://docs.rubixkube.ai/cli/usage

Launch the Rubix CLI. The first run prompts for sign-in via device-code flow.

```bash
rubix
```

--------------------------------

### Enable Required GCP APIs

Source: https://docs.rubixkube.ai/environments/gcp

Manually enable essential Google Cloud APIs if they were not enabled during the initial installer run.

```bash
gcloud services enable compute.googleapis.com container.googleapis.com \
  sqladmin.googleapis.com monitoring.googleapis.com logging.googleapis.com \
  run.googleapis.com cloudfunctions.googleapis.com
```

--------------------------------

### Uninstall Rubix CLI Globally

Source: https://docs.rubixkube.ai/cli/installation

Command to remove the globally installed Rubix CLI. Local authentication credentials are not affected.

```bash
npm uninstall -g @rubixkube/rubix
```

--------------------------------

### Example Conversation Flow for Pod Troubleshooting

Source: https://docs.rubixkube.ai/tutorials/chat-troubleshooting

Demonstrates a natural language conversation to identify and analyze failing pods. The agent maintains context, allowing follow-up questions without repeating pod names.

```text
You: "Show me all failing pods in rubixkube-tutorials"
Agent: [Lists 3 failing pods]

You: "Tell me about the memory one"
      ↑ Agent knows you mean memory-hog-demo

Agent: [Provides OOMKilled analysis]

You: "What about the crash-loop one?"
      ↑ Still in same namespace context

Agent: [Analyzes crash-loop-demo]

You: "And the image pull issue?"
      ↑ Agent infers broken-image-demo

Agent: [Explains ImagePullBackOff]
```

--------------------------------

### Pod Optimization Planning Queries

Source: https://docs.rubixkube.ai/tutorials/chat-cost-analysis

Ask for specific recommendations on how to reduce costs for identified pods, including optimal CPU and memory limits based on application type.

```text
"How can I reduce costs for [pod-name]?"
"What's a good CPU/memory limit for [application-type]?"
"Compare my allocation to best practices"
```

--------------------------------

### Pod Failure Troubleshooting Queries

Source: https://docs.rubixkube.ai/tutorials/chat-troubleshooting

Diagnose why a specific pod is failing, get error messages, and root cause analysis.

```shell
"Why is [pod-name] failing?"
"What's wrong with [pod-name]?"
"Diagnose [pod-name]"
"Show me errors for [pod-name]"
```

--------------------------------

### Investigate a Specific Resource

Source: https://docs.rubixkube.ai/cli/examples

Ask a specific question about a resource to get a single-turn, cited answer directly back to your shell.

```bash
rubix chat --prompt "Why is pod api-gateway-7d4b9c8f6d-abc12 in prod-eks restarting?"
```

--------------------------------

### Traditional Kubernetes Commands

Source: https://docs.rubixkube.ai/tutorials/talk-to-infra

This demonstrates the series of kubectl commands typically required to gather information about a specific service, contrasting with a single natural language query.

```bash
kubectl get pods -n production
kubectl describe pod api-gateway-xyz
kubectl logs api-gateway-xyz
kubectl top pod api-gateway-xyz
# five more commands, then stitch it together in your head
```

--------------------------------

### Pre-Deploy Validation Query

Source: https://docs.rubixkube.ai/tutorials/chat-advanced

Upload your deployment YAML and use these prompts to validate changes and assess deployment risks before applying them.

```bash
[Upload deployment.yaml]
"Validate this"
"What's the risk?"
```

--------------------------------

### Create Query Templates for Daily Use

Source: https://docs.rubixkube.ai/tutorials/chat-advanced

Save frequently used queries as templates for quick access. Paste and run these templates daily for routine checks.

```text
Daily health:    "Cluster status + incidents?"
Pre-deploy:      "Validate [service] for deploy"
Post-deploy:     "How's [service] after deploy?"
Triage:          "Show HIGH severity issues"
```

--------------------------------

### Post-Optimization Verification Queries

Source: https://docs.rubixkube.ai/tutorials/chat-cost-analysis

After applying changes, use these queries to verify resource usage, confirm cost reductions, and check for any performance issues.

```text
"Show me current resource usage for [pod-name]"
"Did we reduce costs successfully?"
"What's our new monthly estimate?"
```

--------------------------------

### Troubleshooting OOMKilled Pod Logs

Source: https://docs.rubixkube.ai/tutorials/chat-troubleshooting

Example log output from a pod experiencing memory issues, indicating high memory consumption.

```text
dispatching hogs: 0 cpu, 0 io, 1 vm, 0 hdd
```