Pillar guide
Claude Kit: What It Is and How to Build Yours
Everything that belongs in a Claude kit — the apps, the agentic coding setup, Skills, MCP servers and prompts — with links to the best free resources and the leading ready-made kit.
Last updated · By Claudekit Team
What is a Claude kit?
A Claude kit is the toolbox you assemble around Claude so it can do real work for you: the surface you use it from, the context it starts with, the reusable instructions it can call on and the systems it can reach. Two developers using the same Claude model can get very different results — the difference is almost always the kit.
Some people buy a ready-made kit; most combine free building blocks. This page covers both, and every section links into the Claudekit library so you can go deeper.
Claudekit.cc is now Agentkit.best
If you searched for “Claudekit.cc”, you were probably looking for the commercial starter kit that used to live at claudekit.cc. It is now Agentkit.best (formerly Claudekit.cc), at agentkit.best. It bundles pre-built Claude Code agents, Skills, commands and workflows so you can skip most of the setup below. Our Agentkit.best guide covers what you get, the price and who should skip it.
Check out Agentkit.best (affiliate link — see our disclosure). Claudekit.ai is a separate, independent library and is not the same product.
The building blocks
1. Where you use Claude
Claude.ai (web, desktop and mobile) is best for chat, research and Projects with shared files. Claude Code is the agentic coding tool that works directly in your terminal and repository. The Claude API is for building Claude into your own products. See the guides for setup walkthroughs.
2. Context: project memory
Claude Code reads a CLAUDE.md file for project instructions — your stack, conventions and the
commands to build and test. A short, accurate memory file is the cheapest upgrade any kit can get.
3. Skills
Skills are folders of instructions, scripts and resources that Claude loads only when a task needs them. They turn a long prompt you keep re-pasting into a reusable capability. Browse the Skills directory.
4. MCP servers
The Model Context Protocol (MCP) is an open standard for connecting Claude to tools and data — GitHub, databases, browsers, docs and internal APIs. Add only the servers you need. Browse MCP servers.
5. Prompts
Even with Skills, a good prompt library pays off for one-off tasks. Keep the ones that work, note the model they were tested on, and retire them when models change. Browse the prompt library.
6. Agents and automation
Claude Code supports subagents, hooks and custom slash commands for multi-step workflows. Starter kits mostly package these pieces. Compare options on the comparison pages.
Build your own kit in 6 steps
- Pick your surface: Claude.ai for chat and Projects, Claude Code for agentic coding in the terminal, or the Claude API for your own apps.
- Write a project memory file (CLAUDE.md for Claude Code) with your stack, conventions and commands so every session starts with context.
- Add a few Skills for repeat workflows — code review, release notes, docs — instead of re-pasting long prompts.
- Connect the MCP servers you actually need (GitHub, your database, docs search) and nothing more; each one widens what Claude can touch.
- Save the prompts that work into a shared library with the model they were tested on.
- Review monthly: remove stale skills and servers, update prompts for new models, and check the news for breaking changes.
Starter picks from the library
Start with these guides
See all →AI Agent Development Cost (2026): Build, Run and Maintain
What a custom AI agent costs in 2026: real build prices from freelancers, fixed-price services and agencies, plus how to estimate token and upkeep costs.
AI Agents for Small Business: 12 Use Cases, DIY or Hire
Twelve narrow jobs an AI agent can do in a small business, each with its trigger, tools, a feature count and whether you can build it yourself or should hire.
Claude Code Pricing (2026): Plans, API Costs and Limits
Claude Code pricing explained: which Claude plans include it, Pro vs Max vs Team seats, API token rates, usage limits, and how to track spend with /usage.
Popular Skills
See all →Superpowers: Agentic Skills Framework for Claude Code
Superpowers by obra is an open-source skills library that adds brainstorming, planning, TDD, systematic debugging and review workflows to Claude Code.
algorithmic-art Skill: Generative Art with p5.js
Anthropic's algorithmic-art skill has Claude create generative p5.js art with seeded randomness, flow fields and particles, plus an interactive viewer. Usage.
canvas-design Skill: Posters and Visual Art with Claude
Anthropic's canvas-design skill has Claude write a design philosophy, then express it as an original poster or artwork in PNG or PDF. How it works and install.
MCP servers worth installing
See all →Apify MCP Server: Run Web Scrapers from Claude
Apify's official MCP server lets Claude find and run Actors from Apify Store to scrape sites, maps and social media. Claude Code plugin, OAuth, caveats.
Atlassian MCP Server: Jira and Confluence in Claude
Atlassian's official remote MCP server connects Claude to Jira, Confluence, JSM, Bitbucket, Compass and Loom. Claude Code setup, admin controls, alternatives.
AWS MCP Servers: Official AWS Tools for Claude Code
awslabs/mcp is AWS's open-source suite of MCP servers for docs, APIs, IaC, pricing, databases and more. Claude Code commands, picks to start with, caveats.
Ready-made kits
See all →Agentkit.best (formerly Claudekit.cc): Claude Code Kit Review
Claudekit.cc changed to Agentkit.best: a paid kit of skills, subagents and slash-command workflows for Claude Code and other agents. What it includes, pros and cons.
Claude Code Templates: Agents, Commands, Hooks, MCPs
Claude Code Templates (aitmpl.com) is an npx CLI and catalog to install agents, commands, hooks, settings, MCPs and skills, plus analytics. Uses, caveats.
Comparisons
See all →Claude Pro vs Max (2026): Which Plan Is Worth It?
Claude Pro vs Max compared: prices, usage multipliers, Claude Code limits, included API credits and a simple test to decide if Pro is worth it or Max pays off.
Agentkit.best (formerly Claudekit.cc) vs DIY and Free Kits
Agentkit.best vs a DIY Claude kit, Superpowers, carlrannaberg's claudekit and official plugins: what each includes, what it costs you, and which one fits your team.
Claude Code vs Codex: Which Coding Agent Fits You?
Claude Code vs OpenAI Codex compared fairly: CLI, IDE and cloud surfaces, safety controls, AGENTS.md vs CLAUDE.md, extensibility, CI, and when to pick each one.
Frequently asked questions
What is a Claude kit?
A Claude kit is the set of tools, configuration and reusable assets you use to work with Claude effectively: usually Claude Code, a project memory file, Skills, MCP servers, prompt templates and sometimes a ready-made starter kit of agents and workflows.
What happened to Claudekit.cc?
Claudekit.cc, the commercial starter kit, changed its name to Agentkit.best and now lives at agentkit.best. Claudekit.ai is a separate, independent community resource library and is not the same product.
Is Claudekit.ai made by Anthropic?
No. Claudekit.ai is an independent community project. It is not affiliated with, endorsed by, or sponsored by Anthropic. Claude is a trademark of Anthropic, PBC.
Do I need a paid kit to use Claude Code well?
No. Everything a kit contains can be assembled from free and open-source pieces: Skills, MCP servers and prompts listed in this library. A paid kit mainly saves setup time by bundling a curated, pre-configured set.
What is the difference between a Skill and an MCP server?
A Skill packages instructions, scripts and resources that Claude loads when a task calls for them. An MCP server is a separate process that exposes tools and data to Claude over the Model Context Protocol. Skills teach Claude how to do something; MCP servers give it access to systems.