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

54,220 results · page 451 of 1,808

  • pdf-ingestionSkillDocs & knowledge

    Lets your agent read PDFs efficiently while keeping structure and fitting within context limits.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • planner-spec-expandSkillMedia

    Lets your agent turn a short product brief into a full spec with design language, acceptance criteria, and an ordered feature list.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • pr-from-diffSkillDev tools

    Helps your agent write a pull request description that reviewers can approve quickly.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • progress-reading-protocolSkillMonitoring & ops

    Has your agent check recent git history and a progress file to reconstruct project state before starting new work.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • prompt-cachingSkillAI & models

    Lets your agent cache unchanging prompt parts so long loops cost less per turn.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • read-the-traceSkillDev tools

    Lets your agent find the real cause of a crash from a stack trace instead of just matching the error type.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • readme-auditSkillDev tools

    Lets your agent check if a project's README has everything a stranger needs to run it.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • rebase-safelySkillDev tools

    Helps your agent rewrite Git history with rebase or squash while protecting work and shared branches.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • reduce-nestingSkillDev tools

    Flattens deeply nested code by converting nested conditionals into readable early-return style.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • revert-surgicalSkillDev tools

    Lets your agent undo a specific bad commit or change without affecting unrelated work.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • schema-diffSkillAI & models

    Lets your agent compare two database schema versions and highlight risky changes before migrations.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • secret-scanSkillFiles & storage

    Lets your agent scan files for hardcoded secrets, keys, and tokens before you commit them.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • self-eval-biasSkillAI & models

    Lets your agent catch and stop its own habit of praising its work instead of reviewing it critically.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • shift-notesSkillFiles & storage

    Lets your agent write and read a handoff notes file so a new session can continue where the last one stopped.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • spec-firstSkillProductivity

    Makes your agent write a goal spec file before starting a task so it stays on track.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • sprint-contractSkillDev tools

    Lets your agent agree with itself on a clear, testable definition of done before writing any code.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • sql-reviewSkillDatabases & data

    Lets your agent review SQL and ORM queries for correctness, safety, and performance before they run.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • structured-outputSkillAI & models

    Helps your agent get reliable, schema-validated JSON responses from AI models with retry on parse failures.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • subagent-fanoutSkillAI & models

    Lets your agent split a big job into independent pieces and run them in parallel on separate fresh agents.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • suggest-next-featuresSkillFiles & storage

    Lets your agent read your git history and progress notes to draft new feature suggestions in a separate file.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • tool-restraintSkillAI & models

    Guides your agent design toward fewer tools so the agent has less to juggle and fails less often.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • using-loopkitSkillDev tools

    Lets your agent discover and load 49 skills available in Loopkit-enabled projects before responding.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • write-failing-test-firstSkillDev tools

    Makes your agent write a failing test that reproduces a bug before fixing it.

    Ready to connect★ 757

    github.com/archive228/loopkit755 stars

    View details
  • neoSkillWeb & browsing

    Lets your agent open websites, read pages, click and fill forms, and call website APIs through Chrome.

    Ready to connect★ 753

    github.com/4ier/neo751 stars

    View details
  • ad-adrSkillDocs & knowledge

    Lets your agent draft a new Architecture Decision Record file from your conversation, ready for your review.

    Ready to connect★ 750

    github.com/alexandremendoncaalvaro/corridorkey-runtime743 stars

    View details
  • ad-architectureSkillDocs & knowledge

    Lets your agent generate or audit an ARCHITECTURE.md for a codebase by scanning the code and pre-filling patterns.

    Ready to connect★ 750

    github.com/alexandremendoncaalvaro/corridorkey-runtime743 stars

    View details
  • ad-archiveSkillFiles & storage

    Lets your agent find finished task, spec, PRD, and ADR files and remove them from the working tree using git rm.

    Ready to connect★ 750

    github.com/alexandremendoncaalvaro/corridorkey-runtime743 stars

    View details
  • ad-auditSkillDev tools

    Lets your agent run a thorough rules-based audit of code or work, giving a pass/fail verdict for every rule.

    Ready to connect★ 750

    github.com/alexandremendoncaalvaro/corridorkey-runtime743 stars

    View details
  • ad-bootstrapSkillAI & models

    Lets your agent generate an AGENTS.md guide for your repo by scanning the code and filling in what it finds.

    Ready to connect★ 750

    github.com/alexandremendoncaalvaro/corridorkey-runtime743 stars

    View details
  • ad-commitSkillCommunication

    Guides your agent to split changes into clean, single-purpose git commits with conventional messages and signed-off attribution.

    Ready to connect★ 750

    github.com/alexandremendoncaalvaro/corridorkey-runtime743 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.

53,788 of the 54,220 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