Skip to main content
skill CX Codex

digitalocean-compute — Claude Code skill: pick and provision

Tells your agent which DigitalOcean compute to use for a job: Droplets, App Platform, Functions, DOKS or GPUs. Handy when you are deciding where an app or container should run.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add bobmatnyc/claude-mpm-skills --skill digitalocean-compute -a claude-code
Open on GitHub

What digitalocean-compute does

The skill lays out six compute options on DigitalOcean and sorts them by workload. Droplets give you full OS and network control. App Platform builds and runs straight from a Git repo or a container image. Functions cover event-driven jobs and small APIs. DOKS is managed Kubernetes for multi-service setups. GPU Droplets and Bare Metal GPUs are aimed at AI and ML work.

Each option comes with a short checklist. Droplets: region and plan, OS image or snapshot, SSH keys, Volumes for block storage, firewalls and VPC. App Platform: connect a repo or registry, set build and run commands, add environment variables and secrets, point a custom domain with HTTPS. Functions: create a namespace, deploy with doctl or the Control Panel, wire triggers, watch logs. DOKS: cluster and node pools, save kubeconfig, apply manifests, pull images from Container Registry.

A selection table maps workload type to service, so the agent can justify a pick instead of guessing. The skill is not user-invocable — it fires on its own when the agent touches version control, branches or pull requests. Related skills for registry, storage, Kubernetes and Docker exist in the same repo, but this one stands alone.

What it helps with

  • choose between Droplets, App Platform and Functions for a given workload
  • deploy an app from a Git repo without hand-building a server
  • spin up managed Kubernetes and save the kubeconfig
  • plan GPU work on Droplets or Bare Metal GPUs
  • set up SSH keys, Volumes and VPC for a VM

Example prompts for the agent

  • Pick a DigitalOcean compute option for a container app with three services
  • Functions or App Platform for an event-driven API?
  • Walk me through creating a DOKS cluster and saving kubeconfig

How to install digitalocean-compute

1Skills CLI (Vercel)

npx skills add bobmatnyc/claude-mpm-skills --skill digitalocean-compute -a claude-code

2Manually: put the files in place

~/.claude/skills/digitalocean-compute/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/digitalocean-compute/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about digitalocean-compute

How do I install digitalocean-compute in Claude?

Run npx skills add bobmatnyc/claude-mpm-skills --skill digitalocean-compute -a claude-code. The skill file then lives at ~/.claude/skills/digitalocean-compute/SKILL.md.

What does digitalocean-compute do?

It covers DigitalOcean compute services — Droplets, App Platform, Functions, DOKS, GPU Droplets and Bare Metal GPUs — and helps pick one for the task at hand. From there it walks through region, size, image, networking and deploy steps.

Can I invoke this skill manually?

No. It sets user-invocable: false and disable-model-invocation: true, so it loads on its own when the agent works with branches, version control or pull requests.

Source text

SKILL.md · MIT
---
name: digitalocean-compute
description: DigitalOcean compute services covering Droplets, App Platform, Functions, Kubernetes (DOKS), GPU Droplets, and Bare Metal GPUs. Use when selecting or provisioning compute for applications, containers, or serverless workloads.
user-invocable: false
disable-model-invocation: true
progressive_disclosure:
  entry_point:
    summary: "DigitalOcean compute services covering Droplets, App Platform, Functions, Kubernetes (DOKS), GPU Droplets, and Bare Metal GPUs. Use when selecting or provisioning compute for applications, containe..."
    when_to_use: "When working with version control, branches, or pull requests."
    quick_start: "1. Review the core concepts below. 2. Apply patterns to your use case. 3. Follow best practices for implementation."
---
# DigitalOcean Compute Skill

---
progressive_disclosure:
  entry_point:
    summary: "Compute on DigitalOcean: Droplets, App Platform, Functions, Kubernetes, GPU Droplets, Bare Metal GPUs."
    when_to_use:
      - "When choosing a compute model for an application"
      - "When provisioning VMs, PaaS, serverless, or Kubernetes"
      - "When planning GPU-backed workloads"
    quick_start:
      - "Pick Droplets, App Platform, Functions, or Kubernetes"
      - "Select region, size, and image"
      - "Configure networking and access"
      - "Deploy and monitor workloads"
  token_estimate:
    entry: 90-110
    full: 4000-5200
---

## Overview

DigitalOcean compute includes VMs, managed PaaS, serverless functions, and managed Kubernetes, plus GPU offerings for AI workloads.

## Droplets (VMs)

Use Droplets for full OS control and custom networking.

- Choose region and Droplet plan.
- Select an OS image or snapshot.
- Add SSH keys for secure access.
- Attach Volumes for persistent block storage.
- Apply firewalls and VPC configuration.

## App Platform (PaaS)

Use App Platform to deploy from Git or container images with managed build and runtime.

- Connect a Git repository or container registry.
- Define build and run commands or app spec.
- Set environment variables and secrets.
- Configure custom domains and HTTPS.

## Functions (FaaS)

Use Functions for event-driven or lightweight API workloads.

- Create a Functions namespace.
- Deploy functions with doctl or the Control Panel.
- Configure environment variables and triggers.
- Monitor execution and logs.

## Kubernetes (DOKS)

Use managed Kubernetes for multi-service container orchestration.

- Create a cluster and node pools.
- Save kubeconfig for cluster access.
- Deploy workloads with Kubernetes manifests.
- Pull images from Container Registry.

## GPU Compute

Use GPU Droplets or Bare Metal GPUs for AI and ML workloads.

- GPU Droplets: VM-based GPU workloads with flexible sizing.
- Bare Metal GPUs: Dedicated GPU hardware for high performance.

## Compute Selection Guide

- Use **Droplets** for full control and custom configurations.
- Use **App Platform** for managed app deployments from Git or containers.
- Use **Functions** for event-driven workflows and micro APIs.
- Use **Kubernetes** for multi-service container orchestration.
- Use **GPU Droplets** or **Bare Metal GPUs** for AI or GPU-heavy workloads.

## Complementary Skills

When using this skill, consider these related skills (if deployed):

- **digitalocean-containers-images**: Registry and image workflows.
- **digitalocean-storage**: Spaces, Volumes, and backups.
- **kubernetes**: Cluster operations and troubleshooting.
- **docker**: Container build and runtime patterns.

*Note: Complementary skills are optional. This skill is fully functional without them.*

## Resources

**DigitalOcean Docs**:
- Compute: https://docs.digitalocean.com/products/compute/
- Droplets: https://docs.digitalocean.com/products/droplets/
- App Platform: https://docs.digitalocean.com/products/app-platform/
- Functions: https://docs.digitalocean.com/products/functions/
- Kubernetes: https://docs.digitalocean.com/products/kubernetes/
- GPU Droplets: https://docs.digitalocean.com/products/gpu-droplets/
- Bare Metal GPUs: https://docs.digitalocean.com/products/bare-metal-gpus/

© bobmatnyc · MIT · bobmatnyc/claude-mpm-skills

Similar skills

Everything for DevOps and infrastructure →

From the same set65

hono-middleware A skill covering Hono middleware: writing your own, composing them, the built-in modules and the order they run in. Useful when a Hono app needs auth, CORS or logging. CX Codex +1 431installs golang-cli-cobra-viper A skill for building Go command-line tools: command structure with Cobra and configuration from flags, environment variables and files with Viper. CX Codex +1 425installs dependency-audit A skill that walks through dependency cleanup: outdated packages, security vulnerabilities and libraries nothing imports anymore. CX Codex +1 422installs wordpress-plugin-fundamentals A skill for building WordPress plugins on PHP 8.3+ with OOP, hooks, database access and the Settings API. Useful when the agent has to scaffold a plugin instead of guessing at structure. CX Codex +1 421installs wordpress-testing-qa A skill for testing WordPress plugins and themes: PHPUnit integration, WP_Mock, PHPCS and CI/CD. Useful when your test setup needs structure. CX Codex +1 413installs golang-observability-opentelemetry A skill for instrumenting Go services: OpenTelemetry for distributed tracing, Prometheus for metrics, slog for structured logging. CX Codex +1 409installs digitalocean-overview This skill maps DigitalOcean accounts, projects, services, and tooling so you can plan a setup through the Control Panel, doctl, REST API, or Terraform. CX Codex +1 408installs hono-validation A skill for validating Hono requests: JSON bodies, forms, query params, headers and cookies. Useful when you want schema types inside the handler. CX Codex +1 408installs

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor