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 450 of 1,808
- View details
wordpress-abilitySkillDev tools
Lets your agent add or modify WordPress abilities when working on WordPress projects.
Ready to connect★ 766
- View details
wordpress-complianceSkillDev tools
Checks that your WordPress code edits follow WordPress.org coding standards for templates, hooks, and sanitization.
Ready to connect★ 766
- View details
wp-rocket-architectureSkillDev tools
Guides your agent when editing WP Rocket's internal structure like services, providers, container wiring, or core caching code.
Ready to connect★ 766
- View details
wp-rocket-frontend-architectureSkillDev tools
Guides your agent when editing WP Rocket's admin JavaScript, CSS, or HTML/Twig templates.
Ready to connect★ 766
- View details
directorSkillMedia
Lets your agent produce full videos from idea to delivery, including scripts, storyboards, and asset generation.
Ready to connect★ 764
- View details
codemanSkillCommunication
Lets your agent manage worker sessions: start them, send prompts, wait for results, read output, and clean up.
Ready to connect★ 755
- View details
a11y-passSkillDev tools
Lets your agent check AI-built UI components for accessibility failures.
Ready to connect★ 757
- View details
active-memory-reminderSkillDocs & knowledge
Lets your agent reload saved decisions and progress from past sessions before starting new work.
Ready to connect★ 757
- View details
adversarial-verifySkillDev tools
Lets your agent review code changes against the goal spec while assuming the code is broken.
Ready to connect★ 757
- View details
authz-checkSkillSecurity
Lets your agent check whether an endpoint verifies a user owns the data it touches, not just that they are logged in.
Ready to connect★ 757
- View details
bisect-regressionSkillDev tools
Lets your agent search commit history to find the exact commit that introduced a bug.
Ready to connect★ 757
- View details
broken-window-checkSkillDev tools
Checks whether the last completed feature still works before starting new work, and reopens it if broken.
Ready to connect★ 757
- View details
changelog-from-diffSkillDev tools
Turns a set of commits or a diff into a clean, user-facing changelog entry.
Ready to connect★ 757
- View details
clean-commitsSkillCommunication
Lets your agent rewrite messy work-in-progress changes into clean commits with messages explaining why.
Ready to connect★ 757
- View details
contract-testSkillDev tools
Lets your agent test that two systems like APIs or integrations agree at their shared interface.
Ready to connect★ 757
- View details
coverage-gapsSkillDev tools
Lets your agent find untested code paths that matter instead of just reporting coverage percentages.
Ready to connect★ 757
- View details
decision-recordSkillDev tools
Lets your agent record the reasoning behind technical choices so future sessions know why.
Ready to connect★ 757
- View details
evaluator-calibrationSkillDev tools
Lets your agent keep a reviewer's scoring standards consistent across long runs by calibrating it with example rubrics.
Ready to connect★ 757
- View details
feature-list-jsonSkillDev tools
Lets your agent list every project feature as structured JSON entries with pass status, priority, and edit restrictions.
Ready to connect★ 757
- View details
flaky-hunterSkillDev tools
Lets your agent diagnose and fix tests that pass sometimes and fail other times.
Ready to connect★ 757
- View details
harness-strippingSkillDev tools
Lets your agent remove scaffolding components one at a time and measure the impact to cut unneeded complexity.
Ready to connect★ 757
- View details
hitl-escalateSkillProductivity
Lets your agent pause a stuck automated task and ask a human for help instead of looping forever.
Ready to connect★ 757
- View details
init-script-contractSkillDev tools
Lets your agent write standard project scripts (init, test, stop, reset, serve) that follow a fixed naming contract.
Ready to connect★ 757
- View details
input-validationSkillDev tools
Lets your agent validate and constrain untrusted external input before it enters your code.
Ready to connect★ 757
- View details
kill-dead-codeSkillDev tools
Finds and removes unused or unreachable code from your project safely.
Ready to connect★ 757
- View details
loading-empty-error-statesSkillMedia
Helps your agent design loading, empty, and error states for UI components that display data.
Ready to connect★ 757
- View details
migration-writerSkillDatabases & data
Lets your agent write safe, reversible database schema changes that follow your repo's conventions.
Ready to connect★ 757
- View details
model-routingSkillAI & models
Lets your agent run Plan/Act/Verify loops using a cheap model for execution and stronger models for planning and checking.
Ready to connect★ 757
- View details
my-new-skillSkillAI & models
Template skill placeholder for AI model tasks, with no defined functionality beyond routing triggers.
Ready to connect★ 757
- View details
owasp-reviewSkillSecurity
Checks code changes for common security flaws like injection and broken authentication before merging.
Ready to connect★ 757
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.