Skills.

Give your AI a better way to work.

A skill is a set of written instructions that teaches an AI how to do one job the way it should be done: review a pull request, plan a migration, write the release notes.

Install one here and it travels with your account into Claude, Claude Code, Cursor and every other client you sign in with.

Filter by category

54,680 results · page 583 of 1,823

  • anchor-prior-testSkillSearch

    Lets your agent test whether a candidate term is common enough in LLM training data before adopting it.

    Ready to connect★ 470

    github.com/llm-coding/semantic-anchors463 stars

    View details
  • arc42-documentationSkillDocs & knowledge

    Lets your agent write and extend arc42 architecture documents with built-in traceability rules.

    Ready to connect★ 470

    github.com/llm-coding/semantic-anchors463 stars

    View details
  • flueSkillDocs & knowledge

    Routes your coding agent to version-matched Flue documentation for building and debugging Flue agents and workflows.

    Ready to connect★ 470

    github.com/shadcn-labs/agentcn453 stars

    View details
  • launch-shadcn-registrySkillDev tools

    Lets your agent validate a custom shadcn/ui registry, submit it to component directories via GitHub PRs, and draft social posts.

    Ready to connect★ 470

    github.com/shadcn-labs/agentcn453 stars

    View details
  • mastra-file-agentsSkillFiles & storage

    Refactors your Mastra code-based agents into the file-based per-directory convention.

    Ready to connect★ 470

    github.com/shadcn-labs/agentcn453 stars

    View details
  • semantic-anchor-onboardingSkillAI & models

    Deprecated alias that forwards to the semantic-contracts-installer skill, which you should use directly instead.

    Ready to connect★ 470

    github.com/llm-coding/semantic-anchors463 stars

    View details
  • semantic-anchor-translatorSkillAI & models

    Lets your agent turn wordy descriptions into precise established terms and suggest relevant terminology.

    Ready to connect★ 470

    github.com/llm-coding/semantic-anchors463 stars

    View details
  • semantic-contracts-installerSkillAI & models

    Lets your agent install shared coding rules and guidelines into its persistent context across popular coding tools.

    Ready to connect★ 470

    github.com/llm-coding/semantic-anchors463 stars

    View details
  • socratic-code-theory-recoverySkillDocs & knowledge

    Lets your agent build a question tree from an existing codebase to recover its design theory before writing docs.

    Ready to connect★ 470

    github.com/llm-coding/semantic-anchors463 stars

    View details
  • chisleSkillDev tools

    Switches your agent into a terse, minimal coding mode that writes lean code and skips extra features.

    Ready to connect★ 434

    github.com/jaypokale/chisle469 stars

    View details
  • chisle-auditSkillFiles & storage

    Lets your agent audit code and docs to spot over-engineered code and bloated writing in a file, diff, or repo.

    Ready to connect★ 434

    github.com/jaypokale/chisle469 stars

    View details
  • chisle-helpSkillDev tools

    Lets your agent show a quick-reference list of Chisle commands when asked for help.

    Ready to connect★ 434

    github.com/jaypokale/chisle469 stars

    View details
  • chisle-reviewSkillFiles & storage

    Lets your agent review code changes and flag bloat like over-engineering and reinvented standard-library code.

    Ready to connect★ 434

    github.com/jaypokale/chisle469 stars

    View details
  • heige-codex-skin-studioSkillDev tools

    Lets your agent install, apply, pause, or resume custom skins for the HeiGe Codex Desktop app.

    Ready to connect★ 465

    github.com/heigeai/heige-codex-skin-studio469 stars

    View details
  • repo-docsSkillDocs & knowledge

    Lets your agent build and update a Markdown guide explaining how a code repository works.

    Ready to connect★ 469

    github.com/yurunchen/repo-docs-skills466 stars

    View details
  • repo-docs-zhSkillDocs & knowledge

    Lets your agent generate and maintain repository documentation written for Chinese readers while keeping code identifiers intact.

    Ready to connect★ 469

    github.com/yurunchen/repo-docs-skills466 stars

    View details
  • chunking-for-llmsSkillAI & models

    Lets your agent split source code into syntax-safe chunks that fit an LLM context window.

    Ready to connect★ 466

    github.com/xberg-io/tree-sitter-language-pack463 stars

    View details
  • detecting-languagesSkillFiles & storage

    Lets your agent detect which programming language a code file or snippet is written in.

    Ready to connect★ 466

    github.com/xberg-io/tree-sitter-language-pack463 stars

    View details
  • extracting-code-structureSkillFiles & storage

    Lets your agent pull functions, classes, imports, docstrings, and syntax errors out of a source code file.

    Ready to connect★ 466

    github.com/xberg-io/tree-sitter-language-pack463 stars

    View details
  • managing-parsersSkillDev tools

    Lets your agent manage downloaded tree-sitter parsers: prefetch for offline use, list, inspect, or clean the cache.

    Ready to connect★ 466

    github.com/xberg-io/tree-sitter-language-pack463 stars

    View details
  • parsing-sourceSkillFiles & storage

    Lets your agent convert source code files into a syntax tree for inspection or analysis.

    Ready to connect★ 466

    github.com/xberg-io/tree-sitter-language-pack463 stars

    View details
  • tree-sitter-language-packSkillAI & models

    Lets your agent parse code in 371 languages to extract structure, symbols, imports, and syntax diagnostics.

    Ready to connect★ 466

    github.com/xberg-io/tree-sitter-language-pack463 stars

    View details
  • claude-design-cardSkillWeb & browsing

    Lets your agent turn any text or URL into styled HTML info cards and save them as PNG screenshots.

    Ready to connect★ 465

    github.com/geekjourneyx/claude-design-card465 stars

    View details
  • easyeda-agentSkillAI & models

    Lets your agent design, inspect, fix, and lay out EasyEDA Pro schematics and PCB boards through a local CLI.

    Ready to connect★ 455

    github.com/zhoushoujianwork/easyeda-agent465 stars

    View details
  • AI SDK DocumentationSkillDocs & knowledge

    Lets your agent look up answers and best practices for the Vercel AI SDK and its packages.

    Ready to connect★ 464

    github.com/malob/nix-config464 stars

    View details
  • deep-research-teamSkillAI & models

    Lets your agent run deep research by coordinating a team of researcher agents to compare and verify sources.

    Ready to connect★ 464

    github.com/malob/nix-config464 stars

    View details
  • icon-lookupSkillCloud & infra

    Lets your agent display Nerd Font icons in terminals that block basic private-use Unicode characters.

    Ready to connect★ 464

    github.com/malob/nix-config464 stars

    View details
  • malo-find-skillsSkillAI & models

    Lets your agent search for and install new skills when you ask how to do something.

    Ready to connect★ 464

    github.com/malob/nix-config464 stars

    View details
  • new-caskSkillDev tools

    Lets your agent create and submit a Homebrew cask that packages a macOS app for installation via Homebrew.

    Ready to connect★ 464

    github.com/malob/nix-config464 stars

    View details
  • search-tipsSkillWeb & browsing

    Guides your agent to run deeper web research across multiple searches, comparisons, and investigations.

    Ready to connect★ 464

    github.com/malob/nix-config464 stars

    View details

What is a skill?

A skill is plain text, usually a SKILL.md file and the scripts it refers to, written for an AI rather than for a person. It carries the steps, the house rules and the examples a good answer needs, so you stop pasting the same briefing into every new chat.

54,249 of the 54,680 skills listed here can be served through ahel today, and they come from public repositories. Each one has its own page with the instructions themselves on it, so you can read what a skill will tell your AI to do before you install it.

Install one and every AI you use gets it

Installing a skill adds it to your gateway and turns it on in the same step. Claude Code surfaces it as a slash command; any client can read the full instructions with the skill_read tool.

Nothing is copied into a project folder. The instructions are served from your account, so the same skill is there in every AI you connect, and turning it off removes it from all of them at once.

See how to connect your AI