VividKit Crystal clear AI coding
VI
Choose Your Setup Get Started with AgentKit Continue with ClaudeKit Migrate from ClaudeKit to AgentKit Clean cutover from ClaudeKit What Is AgentKit? How AgentKit Works AgentKit Kits, Skills, Agents & Hooks AgentKit Runtime Support AgentKit Desktop App AgentKit Configuration AgentKit CLI Quick Reference AgentKit Skills Cheatsheet AgentKit Workflows AgentKit Statusline Update AgentKit CLI & Kits AgentKit Helper Clean reinstall of AgentKit AgentKit Troubleshooting What is ClaudeKit? ClaudeKit CLI Guide Command Finder Commands Cheat Sheet Workflow Recipes CLI Quick Reference Claude Mechanics Permission Modes ClaudeKit Hooks Coexistence ClaudeKit x Codex CK → Other coding tools How ClaudeKit Works Inside ClaudeKit Finding Your Unknowns Session Recovery Fix from Logs UI/UX Guide Accounts Switcher IDE Config Remote Control Codex App System Diagramming UI Review Gate Promotions Support VividKit

Quick Ref / Command

Engineer Kit

/ak:devops

Deploy and operate infrastructure

Deploy and manage infrastructure across Cloudflare, Docker, Google Cloud, Kubernetes, CI/CD, GitOps, and security audits.

01

Classify Platform

02

Select Fit

03

Load Reference

04

Prepare Commands

Rule 01

Choose the platform from workload shape, not habit.

Rule 02

Security basics are part of deployment, not a later cleanup.

Rule 03

Prefer version-controlled infrastructure and reproducible commands.

Rule 04

Optimize for latency, cost, and resource limits based on the selected platform.

Execution Map

Core lanes from input to output, aligned with the command's real execution path.

01

Start

Input, scope, route

  1. 1 Classify Platform Map the request to Cloudflare Workers/Pages/R2/D1, Docker, GCP Cloud Run/GKE/Cloud SQL, Kubernetes, Helm, or GitOps.
  2. 2 Select Fit Use the platform table: edge latency to Workers, large egress to R2, containers to Docker plus Cloud Run/GKE, enterprise clusters to GKE.
02

Work

Agent / skill execution

  1. 3 Load Reference Navigate to the relevant reference guide for Cloudflare, Docker, GCP, Kubernetes, Helm, scripts, or upstream docs before acting.
  2. 4 Prepare Commands Choose concrete wrangler, docker, gcloud, kubectl, or helm commands and keep configuration as version-controlled infrastructure where possible.
03

Verify

Gate, review, validation

  1. 5 Secure Defaults Apply non-root containers, RBAC, secret handling, image scanning, network policies, and least-privilege environment access.
  2. 6 Optimize Runtime Tune multi-stage builds, edge caching, resource limits, right-sized services, and cost-aware storage choices.
04

Close

Report, handoff, artifact

  1. 7 Deploy or Audit Run the chosen deployment, CI/CD, GitOps, or audit workflow and capture the observable platform result.
  2. 8 Report Operations Summarize platform, commands, resources touched, security posture, performance or cost trade-offs, and next operational checks.

Syntax, arguments, subcommands

How to invoke the skill: syntax, positional arguments, shared options, then each subcommand with its own syntax and outcome.

Syntax

/ak:devops [platform] [task]

Arguments

[platform]

Platform

Infrastructure surface to work on, such as Cloudflare, Docker, Google Cloud, Kubernetes, Helm, CI/CD, or GitOps. This selects the operational context; it does not choose a default environment or grant deployment approval.

/ak:devops cloudflare "review this Worker staging config"
[task]

Task

Operational request to design, review, implement, troubleshoot, build, deploy, audit, or prepare rollback steps. Include the account, project, region, environment, resource, allowed effect, and validation boundary when known.

/ak:devops kubernetes "review manifests for RBAC, resources, rollout checks, and rollback commands"

Modes / Routes

cloudflare

Edge/serverless

Research:
Load Cloudflare references
Review:
Check bindings, latency, egress
Validate:
wrangler deploy/dev evidence
docker

Container

Research:
Load Docker guides
Review:
Check non-root, image size, ports
Validate:
docker build/run evidence
gcp

Google Cloud

Research:
Load gcloud services
Review:
Check region, service account, costs
Validate:
gcloud command output
kubernetes

Cluster ops

Research:
Use upstream K8s/Helm docs
Review:
Check RBAC, network policies, resources
Validate:
kubectl/helm state

Sample Prompt

Concrete invocations for each flag, subcommand, mode, or route available in the live workflow.

Cloudflare review Recommended 
/ak:devops cloudflare "review this Worker staging config and prepare deploy, dry-run, and rollback steps"
Use when:
Cloudflare Workers, Pages, R2, D1, KV, or Browser Rendering is central to the infrastructure task.
Expected:
Maps the target to the right Cloudflare surface, loads the relevant references, prepares wrangler-oriented commands, and reports bindings, routes, limits, cost, and rollback evidence.
Docker review 
/ak:devops docker "analyze this Dockerfile and propose build/run improvements"
Use when:
A service needs Dockerfile or Docker Compose review before local development or deployment.
Expected:
Uses Docker references and the analyzer-oriented workflow to check build context, ports, volumes, health checks, multi-stage builds, non-root runtime, and secret handling.
GCP deploy plan 
/ak:devops gcp "prepare a Cloud Run deployment plan for this container image"
Use when:
Google Cloud Run, GKE, Cloud SQL, IAM, region, quota, or gcloud operations define the release risk.
Expected:
Loads Google Cloud and gcloud guidance, resolves project, region, service account, APIs, cost and rollback boundaries, then proposes verifiable Cloud Run or GKE commands.
Kubernetes audit 
/ak:devops kubernetes "review manifests for RBAC, network policies, resources, and Helm rollout risk"
Use when:
A Kubernetes cluster, manifest set, Helm release, RBAC policy, or GitOps workflow needs operational review.
Expected:
Uses upstream Kubernetes and Helm references, checks context and namespace, then reviews RBAC, secrets, network policies, resources, rollout checks, and rollback commands.

Handled Scope

  • Cloudflare Workers, Pages, R2, D1, KV, and browser rendering
  • Dockerfiles, Docker Compose, multi-stage builds, and local development
  • Google Cloud Run, GKE, Cloud SQL, and gcloud operations
  • Kubernetes cluster work with kubectl, Helm, RBAC, and network policies
  • CI/CD, GitOps, multi-region deployments, and security audits

Next