Command Palette

How Command Palette Is Organized

The 304 prompts are organized into 10 categories based on real-world DevOps and SRE workflows:

Category 1: Daily Operations (107 prompts)

Routine status checks, quick lookups, and verifications.

Examples:

  • Pod status and details

  • Deployment listings

  • Service endpoint checks

  • Container log retrieval

  • Node resource inspection

  • Namespace overviews

Why it matters: These are your daily tasks. Having them one click away saves minutes every day.

Category 2: Analytics (57 prompts)

Data analysis, metrics queries, and insights.

Examples:

  • PromQL queries for any metric

  • Resource utilization trends

  • Time-period comparisons

  • Usage reports and summaries

  • Alert rule effectiveness analysis

Why it matters: These are the bread-and-butter operations. Having them one click away saves minutes every day.

Category: Analytics (57 prompts)

Data analysis, metrics queries, and insights from your infrastructure.

What's included:

  • PromQL queries for any metric

  • Resource utilization trends

  • Time-period comparisons

  • Usage reports and summaries

  • Alert rule effectiveness analysis

Why it matters: Capacity planning, performance reviews, and executive reporting all require data. Analytics prompts get you that data fast.

Category: Observability (30 prompts)

Logging, tracing, and monitoring operations.

What's included:

  • Log searches with filters

  • Elasticsearch queries

  • Distributed trace inspection

  • Log-metric correlation

  • Service mesh traffic visualization

Why it matters: When you're investigating performance or debugging, you need to dig into logs and traces. These prompts get you there directly.

Category: Monitoring & Health (27 prompts)

Health checks, status verification, and monitoring configuration.

What's included:

  • Prometheus target health

  • Grafana datasource status

  • Firing alert listings

  • AlertManager configuration review

  • Recording rule inspection

Why it matters: Keeping your monitoring healthy is essential. These prompts help you verify everything is working.

Category: Resource Management (26 prompts)

Managing and organizing infrastructure resources.

What's included:

  • Scaling deployments

  • Resource limit updates

  • ConfigMap and Secret management

  • Unused resource cleanup

  • Namespace organization

Why it matters: Right-sizing and resource hygiene prevent waste and issues. These prompts make resource management easy.

Category: Incident Response (22 prompts)

Operations for investigating and responding to issues quickly.

What's included:

  • Pod crash reason analysis

  • OOMKilled event investigation

  • Recent deployment change review

  • Error spike analysis

  • Failed job inspection

Why it matters: When something goes wrong, you need answers fast. These prompts accelerate initial triage.

Category: Configuration (14 prompts)

Configuration management and inspection.

What's included:

  • ConfigMap content viewing

  • Environment variable inspection

  • Istio configuration review

  • ArgoCD application settings

  • Helm values validation

Why it matters: Configuration issues are a common cause of problems. These prompts help you inspect and verify configs.

Category: Security & Compliance (8 prompts)

Security-related operations and compliance checks.

What's included:

  • RBAC role review

  • Network policy inspection

  • Pod security context audit

  • Service account analysis

  • Secret management verification

Why it matters: Security audits and compliance checks are essential. These prompts produce audit-ready output.

Category: Optimization (7 prompts)

Performance optimization and resource efficiency.

What's included:

  • Over-provisioned resource identification

  • Resource waste analysis

  • Cost optimization opportunities

  • HPA effectiveness review

  • Scheduling efficiency checks

Why it matters: Cloud costs add up. These prompts help you find savings opportunities.

Category: Deployment (6 prompts)

Deployment operations and release management.

What's included:

  • ArgoCD sync triggers

  • Rollout status checks

  • Canary promotion

  • Deployment history

  • Rollback execution

Why it matters: Deployments are high-stakes operations. These prompts give you visibility and control.


Supported Platforms

Ops Mode connects to your entire infrastructure stack:

Infrastructure & Orchestration

Platform
What You Can Do

Kubernetes

Pod management, deployments, services, config maps, resource status, namespace operations

AWS

EC2, EKS, RDS, S3, Lambda, CloudWatch, IAM, VPC, cost analysis, resource management

Helm

Chart management, release status, upgrade operations, values inspection

Monitoring & Observability

Platform
What You Can Do

Prometheus

Metric queries, alert rules, target status, recording rules, PromQL execution

Grafana

Dashboard status, datasource health, alert rules, annotation queries

ELK (Elasticsearch)

Log queries, index health, cluster status, search operations

Deployment & GitOps

Platform
What You Can Do

ArgoCD

Application sync, deployment status, rollback, health checks

Argo Rollouts

Canary deployments, blue-green status, rollout promotion

Istio

Service mesh configuration, traffic policies, gateway status

GitHub

Repository status, PR analysis, workflow runs, commit history

Application Monitoring

Platform
What You Can Do

Sentry

Error tracking, issue analysis, release health, event timelines


Frequently Asked Questions

chevron-rightHow does the Command Palette differ from just asking a question?hashtag

The Command Palette provides pre-optimized prompts for common operations. They're designed to get you accurate results efficiently. Natural language queries work great for custom or complex requests, while the Command Palette is faster for standard operations.

chevron-rightWhat if the Command Palette doesn't have what I need?hashtag

Use natural language. Ops Mode understands a wide range of operational requests beyond the Command Palette. The palette covers the most common workflows, but you're not limited to it.

Last updated