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

54,404 results · page 1069 of 1,814

  • piSkillDev tools

    Lets your agent use the pi skill for programming tasks like debugging, refactoring, testing, and code review.

    Ready to connect★ 108

    github.com/share-skills/pi108 stars

    View details
  • pi-enSkillSearch

    Lets your agent invoke a coding and debugging assistant when it hits dev tasks or gets stuck in retry loops.

    Ready to connect★ 108

    github.com/share-skills/pi108 stars

    View details
  • pi-en-progressiveSkillMedia

    Lets your agent get extra reasoning help on coding tasks or when it gets stuck and keeps failing.

    Ready to connect★ 108

    github.com/share-skills/pi108 stars

    View details
  • pi-progressiveSkillDev tools

    Lets your agent help with coding tasks like debugging, refactoring, testing, and code review.

    Ready to connect★ 108

    github.com/share-skills/pi108 stars

    View details
  • resolve-rebase-conflictsSkillFiles & storage

    Lets your agent resolve conflicts left behind by a stopped git rebase, handling mechanical cases automatically.

    Ready to connect★ 108

    github.com/tbhb/vale-ai-tells83 stars

    View details
  • review-commit-messageSkillCommunication

    Lets your agent review claude skill output like commit messages against the actual code diff.

    Ready to connect★ 108

    github.com/tbhb/vale-ai-tells83 stars

    View details
  • review-pr-descriptionSkillSearch

    Lets your agent review Claude skill drafts and pull request descriptions against the actual code changes for unsupported claims.

    Ready to connect★ 108

    github.com/tbhb/vale-ai-tells83 stars

    View details
  • review-prose-fixSkillDocs & knowledge

    Lets your agent review claude skill changes by checking what a prose fix did to a document.

    Ready to connect★ 108

    github.com/tbhb/vale-ai-tells83 stars

    View details
  • review-squash-messageSkillCommunication

    Lets your agent review claude skill-drafted squash commit messages against every commit they collapse.

    Ready to connect★ 108

    github.com/tbhb/vale-ai-tells83 stars

    View details
  • say-it-straightSkillAI & models

    Lets your agent rewrite prose to be direct, concise, and natural without changing the facts.

    Ready to connect★ 108

    github.com/beefiker/superloopy107 stars

    View details
  • simulate-mipSkillDev tools

    Lets your agent simulate a Moonwell governance proposal against a forked chain before it goes live.

    Ready to connect★ 108

    github.com/moonwell-fi/moonwell-contracts-v2108 stars

    View details
  • skill-best-practicesSkillFiles & storage

    Lets your agent write a best Claude skill by structuring new skill folders and instructions to spec.

    Ready to connect★ 108

    github.com/compozy/kb107 stars

    View details
  • split-rewards-mipSkillDev tools

    Splits an oversized rewards-distribution proposal into smaller chunks that fit Moonbeam's per-transaction gas cap.

    Ready to connect★ 108

    github.com/moonwell-fi/moonwell-contracts-v2108 stars

    View details
  • superloopy-cloneSkillWeb & browsing

    Lets your agent rebuild or replicate an existing website as a governed implementation when you ask it to clone a page.

    Ready to connect★ 108

    github.com/beefiker/superloopy107 stars

    View details
  • superloopy-doctorSkillAI & models

    Helps diagnose Superloopy doctor, install, plugin cache, and setup health problems.

    Ready to connect★ 108

    github.com/beefiker/superloopy107 stars

    View details
  • superloopy-frontendSkillWeb & browsing

    Lets your agent build and edit user interfaces for web, desktop, and mobile apps when explicitly invoked.

    Ready to connect★ 108

    github.com/beefiker/superloopy107 stars

    View details
  • superloopy-loopSkillProductivity

    Lets your agent run tasks through a strict evidence loop that tracks progress, criteria, and completion proof.

    Ready to connect★ 108

    github.com/beefiker/superloopy107 stars

    View details
  • superloopy-researchSkillProductivity

    Lets your agent handle explicitly invoked research tasks and return evidence-backed findings.

    Ready to connect★ 108

    github.com/beefiker/superloopy107 stars

    View details
  • superloopy-slidesSkillWeb & browsing

    Lets your agent create, style, convert, and export HTML slide decks and presentations, including from PowerPoint files.

    Ready to connect★ 108

    github.com/beefiker/superloopy107 stars

    View details
  • systematic-qaSkillDev tools

    Lets your agent run full QA on a project: build, lint, tests, startup checks, and end-to-end workflow verification.

    Ready to connect★ 108

    github.com/compozy/kb107 stars

    View details
  • tlonSkillCommunication

    Lets your agent read Urbit messages, contacts, and channels, and manage groups and profiles.

    Ready to connect★ 108

    github.com/tloncorp/tlon-apps108 stars

    View details
  • tlon-product-guideSkillDev tools

    Lets your agent answer questions about Tlon and Urbit products, including signup, groups, channels, DMs, and bot setup.

    Ready to connect★ 108

    github.com/tloncorp/tlon-apps108 stars

    View details
  • tlon-workflowSkillWeb & browsing

    Guides your agent through fixing a Tlon Messenger bug and opening a tested pull request.

    Ready to connect★ 108

    github.com/tloncorp/tlon-apps108 stars

    View details
  • tlon-workflow-doctorSkillAI & models

    Diagnoses and fixes missing or outdated setup tools needed to run mobile agent workflows in the tlon-apps repo.

    Ready to connect★ 108

    github.com/tloncorp/tlon-apps108 stars

    View details
  • write-prose-fixSkillDocs & knowledge

    Lets your agent reword a document so its lint checks pass when you write Claude skill docs.

    Ready to connect★ 108

    github.com/tbhb/vale-ai-tells83 stars

    View details
  • ade-app-controlSkillMonitoring & ops

    Lets your agent launch a local desktop app, click and type in it, read its logs, and capture screenshots of what it shows.

    Ready to connect★ 107

    github.com/arul28/ade100 stars

    View details
  • ade-autoresearchSkillDocs & knowledge

    Lets your agent find and fix performance bottlenecks in an ADE tab by driving the real UI and measuring CPU, memory, IPC, and render times.

    Ready to connect★ 107

    github.com/arul28/ade100 stars

    View details
  • ade-browserSkillWeb & browsing

    Lets your agent open web pages, click and fill forms, take screenshots, and inspect the DOM in a built-in browser.

    Ready to connect★ 107

    github.com/arul28/ade100 stars

    View details
  • ade-deeplinksSkillFiles & storage

    Lets your agent create and share links that open specific spots in the ADE app like lanes, files, commits, PRs and Linear issues.

    Ready to connect★ 107

    github.com/arul28/ade100 stars

    View details
  • ade-lanes-gitSkillDev tools

    Lets your agent create, sync, commit, push, and rebase git worktree branches (lanes) using ADE commands.

    Ready to connect★ 107

    github.com/arul28/ade100 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.

53,973 of the 54,404 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 workspace 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