—skills · counting…
30,375MCP servers
Bridge the gap between your terminal and cloud infrastructure. Discover skills designed to generate Dockerfiles, audit Kubernetes manifests, and manage deployment pipelines autonomously.
Creates optimized multi-stage Dockerfiles for any language or framework, with lean base images, layer caching and hardened security practices built in.
npx skills add github/awesome-copilot --skill multi-stage-dockerfileWrites and audits Kubernetes manifests — deployments, RBAC, NetworkPolicies, Helm charts, resource limits and storage — and helps debug crashing pods from their logs.
npx skills add jeffallan/claude-skills --skill kubernetes-specialistProduction-ready GitHub Actions workflows for testing, building and pushing Docker images and deploying to Kubernetes, including matrix builds and reusable templates.
npx skills add wshobson/agents --skill github-actions-templatesHashiCorp's official skill for writing and reviewing Terraform HCL that follows their style conventions, file organization and best practices.
npx skills add hashicorp/agent-skills --skill terraform-style-guideEvery card above has its own npx skills add command and a copy button. Installing only requires Node.js; a few skills need an extra tool or API key, noted in their description.
The CLI asks which agents to install for — Claude Code, Cursor and other supported agents — and adds the skill to your project. Add -g to install it for your user account across every project.
Start a new agent session and ask in plain words. The agent loads the skill when your request matches its description. Read the SKILL.md on GitHub before installing anything on a system that matters.
As few as possible. Give the agent read-only cloud credentials by default, scope tokens to a single project or namespace, and require a human approval for anything that deploys or deletes. Never paste production secrets into a prompt — keep them in environment variables or your secret manager.
Only if you hand it credentials and allow the commands. A safe pattern is to let the agent generate manifests, Dockerfiles or Terraform, run kubectl diff or terraform plan, and keep the final apply behind your own confirmation or your CI pipeline. Claude Code asks for permission before running shell commands unless you have allowed them.
Commit what the agent produces — workflows, manifests, Terraform modules — so it goes through pull-request review like any other change. Pin action and image versions, and give pipelines short-lived cloud credentials (for example GitHub OIDC) instead of long-lived keys.
Code review, conventional commits, test-driven development and safe refactoring.
Explore Coding skillsTechnical SEO audits, crawls, structured data and Core Web Vitals fixes.
Explore SEO skillsScrape pages to Markdown, crawl whole sections and drive real browsers.
Explore Scraping skillsSecurity code review, Semgrep scans, secret scanning and dependency audits.
Explore Security skillsBuild, generate and call Model Context Protocol servers and MCP Apps.
Explore MCP skillsThe open skills directory for Claude Code & AI agents
© 2026 findaskill