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 533 of 1,823
- View details
data-statistical-analysisSkillMonitoring & ops
Lets your agent run statistical analysis on workspace data, like trends, outliers, and hypothesis tests.
Ready to connect★ 539
- View details
data-validateSkillDatabases & data
Lets your agent check an analysis for methodology flaws, calculation errors, and bias before you share it.
Ready to connect★ 539
- View details
data-write-querySkillDatabases & data
Lets your agent write optimized SQL queries for PostgreSQL, including multi-join and aggregated queries.
Ready to connect★ 539
- View details
db-mongoSkillDatabases & data
Lets your agent query and explore data in your MongoDB databases.
Ready to connect★ 539
- View details
db-mysqlSkillDatabases & data
Lets your agent run queries against your MySQL or MariaDB databases to explore and audit data.
Ready to connect★ 539
- View details
db-postgresSkillDatabases & data
Lets your agent run queries against your PostgreSQL databases to explore and audit data.
Ready to connect★ 539
- View details
dev-ai-slop-cleanerSkillAI & models
Cleans up AI-generated code by removing unneeded helpers, over-abstractions, generic names, and excess configurability.
Ready to connect★ 539
- View details
dev-askSkillAI & models
Lets your agent ask Claude, Codex, or Gemini for a quick second opinion on a question.
Ready to connect★ 539
- View details
dev-autopilotSkillAI & models
Lets your agent autonomously build working code from an idea by planning, coding, and testing it end to end.
Ready to connect★ 539
- View details
dev-cancelSkillDev tools
Lets your agent cleanly stop an active dev workflow like an autopilot or planning session and report what got done.
Ready to connect★ 539
- View details
dev-ccgSkillProductivity
Lets your agent run the same task through Claude, Codex, and Gemini and combine their answers into one.
Ready to connect★ 539
- View details
dev-configure-notificationsSkillCommunication
Lets your agent set up Telegram, Discord, or Slack notifications for alerts like build failures or finished tasks.
Ready to connect★ 539
- View details
dev-deep-diveSkillDev tools
Lets your agent investigate unclear problems by tracing root causes and then clarifying requirements through follow-up questions.
Ready to connect★ 539
- View details
dev-deep-interviewSkillDev tools
Lets your agent ask probing questions to turn vague requirements into a clear, testable spec.
Ready to connect★ 539
- View details
dev-deepinitSkillFiles & storage
Lets your agent create context files like AGENTS.md that give coding agents full project background from the start.
Ready to connect★ 539
- View details
dev-external-contextSkillDocs & knowledge
Lets your agent fetch SDK references, API docs, and framework guides for multiple topics in parallel.
Ready to connect★ 539
- View details
dev-learnerSkillDev tools
Lets your agent turn repeated patterns from a conversation into reusable skills.
Ready to connect★ 539
- View details
dev-mcp-setupSkillSearch
Sets up workspace integrations like web search, filesystem, GitHub, and Stripe for your agent.
Ready to connect★ 539
- View details
dev-planSkillAI & models
Lets your agent turn vague ideas into structured implementation plans, optionally asking you clarifying questions first.
Ready to connect★ 539
- View details
dev-project-session-managerSkillDev tools
Lets your agent create separate git worktrees per issue or PR so parallel work streams don't collide.
Ready to connect★ 539
- View details
dev-ralphSkillProductivity
Lets your agent keep iterating on a task until it's resolved, with a circuit breaker to stop runaway loops.
Ready to connect★ 539
- View details
dev-ralplanSkillDev tools
Lets your agent weigh decisions through a planner, architect, and critic loop before acting.
Ready to connect★ 539
- View details
dev-releaseSkillDev tools
Lets your agent prepare releases by generating changelogs, bumping versions, and creating tags.
Ready to connect★ 539
- View details
dev-rememberSkillDocs & knowledge
Lets your agent save notes like decisions and gotchas so they last beyond the current conversation.
Ready to connect★ 539
- View details
dev-sciomcSkillDev tools
Lets your agent debug by forming a hypothesis, running experiments, and weighing evidence.
Ready to connect★ 539
- View details
dev-skillifySkillDev tools
Lets your agent create a new custom skill by guiding you through defining its trigger, steps, and outputs.
Ready to connect★ 539
- View details
dev-traceSkillMonitoring & ops
Lets your agent systematically rank competing explanations for an observation using evidence for and against each.
Ready to connect★ 539
- View details
dev-ultraqaSkillDev tools
Lets your agent repeat build, lint, test, and fix cycles up to 5 times until all checks pass.
Ready to connect★ 539
- View details
dev-verifySkillAI & models
Lets your agent verify that a code change actually works before claiming it's done.
Ready to connect★ 539
- View details
dev-visual-verdictSkillDev tools
Lets your agent capture before/after screenshots of UI changes and compare them to spot visual regressions.
Ready to connect★ 539
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.