Find a Claude Code skill for the job.

find-skills

Helps the agent find and install agent skills from the open ecosystem with the skills CLI.

hyperframes-cli

Guides the agent through the HyperFrames CLI to scaffold, check, preview, render and publish video projects.

agent-browser

Drives Chrome through a CLI using accessibility snapshots and element refs for browsing, forms and scraping.

grill-with-docs

Interviews you relentlessly about a plan or design while recording decisions in glossary and ADR files.

media-use

Finds or generates music, sound effects, images, voiceover and cutouts for HyperFrames videos, via HeyGen or free local tools.

improve-codebase-architecture

Makes the agent scan a codebase for shallow modules and present deepening options in an HTML report.

tdd

Guides the agent through test-driven red-green-refactor loops with rules for good tests and seams.

hyperframes-registry

Points the agent to about 450 installable HyperFrames video blocks, such as transitions, titles and charts.

setup-matt-pocock-skills

Scaffolds per-repo config for the engineering skills: issue tracker, triage labels, and domain doc layout.

handoff

Compacts the current conversation into a handoff document saved to a temp folder for another agent.

frontend-design

Directs the agent to build UI with distinctive, brief-specific visual choices instead of templated defaults.

grilling

Interviews the user in rounds about a plan, giving a recommended answer to each question until aligned.

triage

Moves issues and external PRs through triage roles, verifies them, and writes agent-ready briefs.

prototype

Builds a throwaway prototype, a logic walkthrough or UI variants, to answer one design question.

domain-modeling

Sharpens a project's domain language, glossary and ADRs as design decisions are made.

codebase-design

Gives the agent a shared vocabulary and principles for designing deep modules with clean seams.

diagnosing-bugs

Guides feedback-loop-first diagnosis of hard bugs and slow regressions, redacting secrets from output.

vercel-react-best-practices

Applies Vercel's React and Next.js performance rules when writing, reviewing or refactoring code.

code-review

Makes the agent review changes since a fixed point against repo standards and the originating spec.

web-design-guidelines

Reviews UI files against fetched Web Interface Guidelines and reports issues with file and line references.

implement

Implements a spec or tickets, runs tdd and code review, then commits to the current branch.

ask-matt

Routes you to the right skill or workflow for your situation, from idea to shipped code.

wayfinder

Maps a too-large effort as decision tickets on the tracker and resolves them until the route is clear.

to-spec

Turns the current conversation into a spec and publishes it to the issue tracker, with no interview.

research

Makes the agent research a question from primary sources and save cited findings as a Markdown file.

to-tickets

Splits a plan into vertical tracer-bullet tickets with blocking edges and publishes them to the tracker.

remotion-best-practices

Routes Remotion video tasks to the matching guide for creating, previewing, multimedia and interactive editing.

design-taste-frontend

Guides web page and app screen design with researched art directions and a strict anti-template style.

caveman

Makes the agent reply in terse, caveman-style prose to cut output tokens while keeping technical facts.

microsoft-foundry

Guides building, deploying, evaluating and managing Microsoft Foundry agents, models and resources end to end.

document-skills

4 skills

xlsx · docx · pptx · pdf

Creates, edits, cleans, and converts spreadsheets (.xlsx, .csv, .tsv), including formulas and charts. Creates, reads, edits, and converts Word (.docx) and .dotx files, including tracked changes and comments.

example-skills

12 skills

algorithmic-art · brand-guidelines · canvas-design · doc-coauthoring · frontend-design · internal-comms · mcp-builder · skill-creator · slack-gif-creator · theme-factory · web-artifacts-builder · webapp-testing

Creates original p5.js generative art with seeded randomness, flow fields, and adjustable parameters. Applies Anthropic's official brand colors and typography to artifacts that need its look and feel.

claude-api

1 skills

claude-api

Reference for the Claude API and Anthropic SDK covering models, pricing, tool use, caching, and migration.

academy-guide

1 skills

academy-guide

Recommends matching Claude Academy courses and tutorials when users ask how to use Claude or its features.

discernment-nudge

1 skills

discernment-nudge

After a substantive answer, asks 2-3 follow-up questions to help the user check facts and assumptions.

superpowers

Adds composable skills that make the agent brainstorm, plan, write tests first, and review code before finishing work.

ponytail

Makes the agent write only the code a task needs: it reuses existing code and libraries first, and keeps validation, security and tests.

agent-skills

Production-grade engineering skills that make coding agents follow spec, plan, build, test, review, and ship workflows.

taste-skill

Makes the agent design distinct websites and app screens, with a chosen art direction, real credited photos, and visual checks in a real browser.

claude-mem

Captures tool usage in Claude Code sessions, generates semantic summaries, and makes them available to future sessions as persistent memory.

impeccable

Adds commands to steer and review UI design (critique, audit, polish, bolder) plus a rule-based detector for generic AI-style design.

archify

Turns a system, idea, question, or plan into an interactive HTML diagram or one-page explainer that the agent can refine.

Understand-Anything

Analyzes a codebase, wiki, or docs into an interactive knowledge graph you can explore, search, and ask questions about.

diagram-design

Draws editorial diagrams as one self-contained HTML and SVG file in your brand's style, across 45 diagram types.

humanizer

Rewrites AI-generated prose to sound human-written while keeping its meaning, based on Wikipedia's signs of AI writing.

i-have-adhd

Makes the coding assistant lead with the next action, number multi-step tasks, and drop filler, preambles and recaps.

marketingskills

Gives AI coding agents marketing skills for conversion optimization, copywriting, SEO, analytics, and growth engineering.

scientific-agent-skills

Gives AI agents 177 scientific skills for biology, chemistry, medicine, physics, engineering, and data analysis workflows.

agents

Adds 94 plugins of agents, skills, and commands for Python, JavaScript, code review, testing, infrastructure, and security work.

huashu-design

Gives the agent design skills for interactive prototypes, slide decks, timeline animations, infographics, and five-dimension design reviews.

planning-with-files

Keeps task_plan.md, findings.md and progress.md on disk so the agent can resume multi-step work after context loss or /clear.

baoyu-skills

Lets the agent generate Xiaohongshu image cards, infographics and article illustrations, and publish articles to WeChat Official Accounts.

claude-seo

Runs full SEO audits on any website, with parallel specialist agents covering technical SEO, content quality, schema markup, and AI search visibility.

Auto-claude-code-research-in-sleep

A set of research skills that run an automated research workflow in Claude Code, with an independent model reviewing the executor's work.

prompt-master

Writes sharper, token-efficient prompts for many AI tools, adapting each prompt's structure to the tool it targets.

video-shotcraft

Turns your real pages and copy into an editable promo MP4 in Remotion, using tuned motion recipes and beat-synced audio.

claude-skills

Ships 67 specialized skills for languages, frameworks, infra, testing, security and data, plus 9 Jira/Confluence workflow commands.

dev-browser

Lets coding agents control a persistent Chrome browser through short JavaScript scripts, with page snapshots for inspection.