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 618 of 1,823
- View details
claude-printSkillAI & models
Lets your agent send a one-off question to Claude and get its answer as a second opinion.
Ready to connect★ 403
- View details
commit-rulesSkillCommunication
Gives your agent shared rules to follow when writing git commit messages.
Ready to connect★ 403
- View details
commit-staged-pushSkillDev tools
Lets your agent commit staged git changes and push them to the remote in one step.
Ready to connect★ 403
- View details
consult-codexSkillDev tools
Lets your agent ask Codex for a second opinion, brainstorm, or discuss problems in a back-and-forth chat.
Ready to connect★ 403
- View details
consult-oracleSkillWeb & browsing
Lets your agent ask ChatGPT for help on hard problems by automating the ChatGPT website in a browser.
Ready to connect★ 403
- View details
contribute-turboSkillDev tools
Lets your agent contribute improvements back to turbo by filing a GitHub issue in the turbo repo.
Ready to connect★ 403
- View details
create-changelogSkillDev tools
Lets your agent create a CHANGELOG.md file with version history pulled from GitHub releases or git tags.
Ready to connect★ 403
- View details
create-project-skillsSkillFiles & storage
Lets your agent scan a codebase and generate skill files that capture its coding conventions.
Ready to connect★ 403
- View details
create-test-planSkillDev tools
Lets your agent analyze code changes and generate a structured test plan covering basic to adversarial scenarios.
Ready to connect★ 403
- View details
create-threat-modelSkillSecurity
Lets your agent analyze a codebase and write a structured threat model file covering assets, attack surfaces, and risks.
Ready to connect★ 403
- View details
discuss-changeSkillDev tools
Lets your agent interview you with questions to align on a planned change before implementing it.
Ready to connect★ 403
- View details
draft-planSkillFiles & storage
Lets your agent write an implementation plan to a markdown file before you start coding.
Ready to connect★ 403
- View details
evaluate-findingsSkillDev tools
Lets your agent judge code review feedback and pick which suggestions are actually worth applying.
Ready to connect★ 403
- View details
explain-thisSkillDev tools
Lets your agent explain whatever you're pointing at — code, errors, command output, or a report — in plain language.
Ready to connect★ 403
- View details
exploratory-testSkillDev tools
Lets your agent run multi-level exploratory tests of an app, from basic features to adversarial scenarios.
Ready to connect★ 403
- View details
fetch-pr-commentsSkillDev tools
Lets your agent fetch and summarize review comments and unresolved threads on a GitHub pull request.
Ready to connect★ 403
- View details
finalizeSkillDev tools
Lets your agent run a wrap-up workflow that tests, polishes, reviews, and commits finished code changes.
Ready to connect★ 403
- View details
find-dead-codeSkillAI & models
Lets your agent scan a codebase to find unused functions, exports, and other dead code.
Ready to connect★ 403
- View details
github-voiceSkillMedia
Gives your agent shared style rules for writing GitHub pull request comments, titles, descriptions, issues, and proposals.
Ready to connect★ 403
- View details
implement-planSkillFiles & storage
Lets your agent carry out a saved implementation plan by running prep, executing each step, and finalizing.
Ready to connect★ 403
- View details
interpret-feedbackSkillDev tools
Lets your agent analyze reviewer feedback to explain what it means, flag correctness concerns, and clarify ambiguities.
Ready to connect★ 403
- View details
map-codebaseSkillDocs & knowledge
Lets your agent scan a codebase's structure, stack, APIs and dependencies and write an architecture report.
Ready to connect★ 403
- View details
note-improvementSkillDocs & knowledge
Lets your agent save an out-of-scope improvement idea to a backlog so it isn't forgotten.
Ready to connect★ 403
- View details
pick-next-issueSkillDev tools
Lets your agent fetch and rank open GitHub issues by popularity and plan work on the top pick.
Ready to connect★ 403
- View details
polish-codeSkillDev tools
Lets your agent clean up code by formatting, linting, testing, and reviewing it until results are stable.
Ready to connect★ 403
- View details
quick-finalizeSkillDev tools
Lets your agent wrap up a code change quickly: clean it up, run checks, update the changelog, and ship.
Ready to connect★ 403
- View details
refine-planSkillDev tools
Lets your agent repeatedly review and improve a plan until no new issues are found.
Ready to connect★ 403
- View details
reply-to-pr-conversationSkillCommunication
Lets your agent draft and post a single reply to comments on a GitHub pull request conversation.
Ready to connect★ 403
- View details
reply-to-pr-threadsSkillDev tools
Lets your agent draft and post replies to GitHub pull request review threads, skipping already-resolved ones.
Ready to connect★ 403
- View details
resolve-findingsSkillDev tools
Lets your agent apply code fixes from evaluated findings directly or via a planning step.
Ready to connect★ 403
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.