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
55,435 results · page 890 of 1,848
- View details
install-rulesSkillAI & models
Lets your agent save plugin rules into your project so they persist without the plugin loaded.
Ready to connect★ 189
- View details
install-scriptsSkillAI & models
Lets your agent install plugin runner scripts into a project folder so they keep working without the plugin loaded.
Ready to connect★ 189
- View details
instruction-dashboard-tuningSkillAI & models
Lets your agent improve dashboard-building instructions by running test builds with discovery, builder, and reviewer sub-agents.
Ready to connect★ 189
- View details
instruction-tuningSkillFiles & storage
Lets your agent improve its instruction files by running test sub-agents, watching where they go wrong, and fixing the instructions.
Ready to connect★ 189
- View details
issue-analyzeSkillDev tools
Lets your agent analyze GitHub issues and PR review comments to find root causes and plan triage.
Ready to connect★ 189
- View details
keelSkillWeb & browsing
Lets your agent plan and build new software projects end to end, from discovery and spec through design and code.
Ready to connect★ 189
- View details
load-pr-reviewSkillDev tools
Lets your agent load GitHub pull request review comments and analyze, triage, and plan fixes.
Ready to connect★ 189
- View details
merge-prepSkillFiles & storage
Lets your agent analyze a branch before merging, reporting commits, conflicts, and affected files.
Ready to connect★ 189
- View details
necessity-auditSkillDev tools
Lets your agent review a spec and flag over-designed requirements, abstractions, and configs that aren't needed.
Ready to connect★ 189
- View details
op-sessionSkillMonitoring & ops
Lets your agent start a 1Password CLI session so it can fetch secrets without repeated biometric prompts.
Ready to connect★ 189
- View details
plan-reviewSkillFiles & storage
Lets your agent get an adversarial review of a drafted plan before presenting it to the user.
Ready to connect★ 189
- View details
post-dev-recapSkillDev tools
Lets your agent create a guided recap of a just-finished feature, including docs and follow-up questions.
Ready to connect★ 189
- View details
post-dev-testSkillFiles & storage
Lets your agent write missing integration and end-to-end tests and report coverage after a feature is built.
Ready to connect★ 189
- View details
pr-commentSkillDev tools
Lets your agent post review comments and feedback on a GitHub pull request.
Ready to connect★ 189
- View details
pre-pr-auditSkillDev tools
Scores your code changes across five dimensions to check if they're ready to submit as a pull request.
Ready to connect★ 189
- View details
precommitSkillDev tools
Lets your agent run pre-commit checks on your code: auto-fix lint issues, build, and run tests.
Ready to connect★ 189
- View details
precommit-fastSkillDev tools
Runs quick pre-commit checks by auto-fixing lint issues then running your tests before you commit.
Ready to connect★ 189
- View details
project-auditSkillSecurity
Lets your agent add, edit, and sync shared calendar events with a team by writing iCal files.
Ready to connect★ 189
- View details
project-briefSkillDev tools
Lets your agent turn a technical spec into a short executive summary for PMs and CTOs.
Ready to connect★ 189
- View details
publish-reportSkillDocs & knowledge
Lets your agent turn local HTML or Markdown reports into public shareable links.
Ready to connect★ 189
- View details
recap-askSkillDocs & knowledge
Lets your agent answer follow-up questions about an existing recap document.
Ready to connect★ 189
- View details
recap-docSkillDocs & knowledge
Generates a recap document walking through what changed after an AI implements a feature.
Ready to connect★ 189
- View details
remindSkillAI & models
Lets your agent reload forgotten rules or missed steps by pulling the right instructions back into context.
Ready to connect★ 189
- View details
repo-intakeSkillFiles & storage
Lets your agent build a one-time inventory map of a codebase, listing entrypoints, tests, and next steps.
Ready to connect★ 189
- View details
req-analyzeSkillProductivity
Lets your agent break down project requirements into structured documents with stakeholder analysis.
Ready to connect★ 189
- View details
request-trackingSkillDocs & knowledge
Lets your agent look up request statuses, track progress, and manage document references in a knowledge base.
Ready to connect★ 189
- View details
review-specSkillDocs & knowledge
Lets your agent review technical spec documents for completeness, feasibility, risk, and consistency with the code.
Ready to connect★ 189
- View details
risk-assessSkillMonitoring & ops
Lets your agent assess the risk of uncommitted code changes, flagging breaking changes and how widely they could affect the codebase.
Ready to connect★ 189
- View details
safe-removeSkillAI & models
Lets your agent remove agent skills, rules, and scripts while checking for dependencies and cleaning up references.
Ready to connect★ 189
- View details
seek-verdictSkillAI & models
Lets your agent get an independent second opinion to confirm, dismiss, or clarify a code review finding.
Ready to connect★ 189
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.
55,004 of the 55,435 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.