Agent skills
Skills for AI coding agents, each with what it costs to load, how much of it is scripts and what it touches on your machine.
How to read a skill
What it costs to keep
Idle in context, in use when it runs, and up to with its linked files.
Scripts or instructions
Script-led, mixed or model-led: how much the model has to work out itself.
What it can reach
Shell, network, file writes, git and credentials, declared or detected.
Quality, security, repo health
Out of 100, each with what it is made of. Turn a card over to see them.
systematic-debugging
322,961Code Review And Quality · Developer Tools
A four-phase debugging process that finds the root cause before any fix is proposed.
test-driven-development
322,961Code Review And Quality · Developer Tools
Enforces red-green-refactor: a failing test first, minimal code to pass it, and deletion of any code written before its test.
using-git-worktrees
322,961Developer Tools · Devops And Releases
Creates an isolated git workspace for feature work, using your harness's native tool first and git worktree as a fallback.
using-superpowers
322,961Agent Workflows
The bootstrap skill that makes your agent check for a relevant skill before any response, including clarifying questions.
verification-before-completion
322,961Agent Workflows · Code Review And Quality · Developer Tools
Requires fresh command output as evidence before your agent claims work is done, fixed or passing.
writing-skills
322,961Agent Workflows
Applies TDD to skill authoring: baseline an agent without the skill, write the skill, then close loopholes until it complies.
claude-api
198,136Developer Tools
Anthropic's reference skill for building with the Claude API: current model IDs and pricing, SDK patterns in seven languages and cURL, Managed Agents, migrations and evals.
frontend-design
198,136Design And Visual Art · Developer Tools · Frontend And UI Engineering
Design guidance that pushes Claude toward distinctive, intentional UI instead of templated defaults, from palette and type to layout and copy.
mcp-builder
198,136Developer Tools
Anthropic's skill for building MCP servers: research the API, implement tools in TypeScript or Python, test with the Inspector, then measure the server with an LLM evaluation harness.
skill-creator
198,136Agent Workflows · Developer Tools
Anthropic's skill for making skills: Claude drafts a SKILL.md, runs test prompts with and without it, shows the results in a review viewer and tunes the description that triggers it.
web-artifacts-builder
198,136Developer Tools · Frontend And UI Engineering
Anthropic's skill for complex claude.ai artifacts: scaffold a React, TypeScript, Tailwind and shadcn/ui project with one script, then bundle it into a single HTML file with another.
webapp-testing
198,136Code Review And Quality · Developer Tools
Anthropic's skill for testing local web apps: Claude writes native Python Playwright scripts, starts the dev servers with a helper, then inspects screenshots, the DOM and console logs.
browser-testing-with-devtools
119,431Code Review And Quality
Give an agent live browser data through Chrome DevTools MCP: DOM, console, network and performance traces, with safety rules.
code-review-and-quality
119,431Code Review And Quality
Review every change on five axes with severity labels, change sizing rules and a checklist before it merges.
code-simplification
119,431Code Review And Quality
Simplify working code for clarity while keeping behavior exactly the same, one tested change at a time.
constraint-driven-development
119,431Code Review And Quality
Write down a project's quality bar in CONSTRAINTS.md with default thresholds, wire checks to the dev loop and catch agents that weaken it.
context-engineering
119,431Agent Workflows
Control what an AI coding agent sees with rules files, a context hierarchy, packing strategies and budget management.
debugging-and-error-recovery
119,431Code Review And Quality
Debug in six steps: stop the line, reproduce, localize, reduce, fix the root cause and guard with a regression test.
doubt-driven-development
119,431Code Review And Quality
Cross-examine every non-trivial decision with a fresh-context adversarial reviewer before it stands.
test-driven-development
119,431Code Review And Quality
Write the failing test first, make it pass with minimal code, then refactor, and reproduce every bug with a test.
using-agent-skills
119,431Agent Workflows
Map incoming work to the right agent skill and follow six operating behaviors that apply across all of them.
Neat Freak
22,434Agent Workflows
End-of-session cleanup that aligns project docs, CLAUDE.md or AGENTS.md rule files and agent memory with what the code and running services do today.
SkillOpt Sleep for Claude Code
19,174Agent Workflows
Gives Claude Code a sleep cycle: it reads past sessions, replays recurring tasks and stages CLAUDE.md and SKILL.md edits that beat a held-out score.
SkillOpt Sleep for Codex
19,174Agent Workflows
Gives Codex a sleep cycle: it reviews past sessions, replays recurring tasks and stages skill and CLAUDE.md edits that pass a held-out score gate.