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
55,055 results · page 678 of 1,836
- View details
Gauntlet GraphSkillSearch
Lets your agent build, search, and query a knowledge graph of your codebase.
★ 335
- View details
gauntlet-curateSkillAI & models
Lets your agent audit a coding problem bank for gaps and draft new YAML entries.
Ready to connect★ 335
- View details
gemini-delegationSkillProductivity
Lets your agent hand tasks to Google's Gemini CLI for analysis, summarization, and very large-context jobs.
Ready to connect★ 335
- View details
Git CatchupSkillAI & models
Lets your agent summarize recent git commits since a baseline, explaining what changed, why, and what to watch for.
★ 335
- View details
git-platformSkillCloud & infra
Lets your agent detect whether a repo uses GitHub, GitLab, or Bitbucket and pick the right commands for issues and PRs.
Ready to connect★ 335
- View details
glm-delegationSkillProductivity
Lets your agent hand tasks to the Z.ai GLM model by rerouting the Claude Code tool to that service, including a 1M-token context mode.
Ready to connect★ 335
- View details
graduated-implementationSkillAI & models
Guides your agent to build features in small increments, only increasing scope once you understand each step.
Ready to connect★ 335
- View details
graph-buildSkillSearch
Lets your agent build or update a code knowledge graph used for search and impact analysis.
Ready to connect★ 335
- View details
graph-searchSkillSearch
Lets your agent search a code knowledge graph for functions, classes, or types by name or path.
Ready to connect★ 335
- View details
hook-authoringSkillSecurity
Guides your agent in writing secure, well-structured hooks that validate tool inputs and enforce rules before actions run.
Ready to connect★ 335
- View details
hook-scope-guideSkillAI & models
Guides your agent to pick the right scope (plugin, project, or global) when writing a hook.
Ready to connect★ 335
- View details
- View details
Hooks EvalSkillAI & models
Lets your agent check a plugin's hooks for quality and compliance issues.
★ 335
- View details
hooks-evalSkillSecurity
Lets your agent audit hook configurations for security flaws, performance issues, and SDK compliance with a scoring framework.
Ready to connect★ 335
- View details
Install WatchdogSkillAI & models
Lets your agent install a watchdog daemon that automatically relaunches sessions.
★ 335
- View details
install-watchdogSkillAI & models
Lets your agent set up a watchdog daemon that automatically restarts a service using launchd or systemd.
Ready to connect★ 335
- View details
justifySkillAI & models
Reviews code changes for added bias and compliance with defined rules before you merge them.
Ready to connect★ 335
- View details
Karpathy CheckSkillAI & models
Checks your AI agent's plan against four coding principles like simplicity and surgical edits before it writes code.
★ 335
- View details
karpathy-principlesSkillAI & models
Makes your agent check its plan for simplicity, surgical edits, and verifiable goals before writing code.
Ready to connect★ 335
- View details
knowledge-intakeSkillDocs & knowledge
Lets your agent turn shared articles, papers, and docs into scored, organized knowledge entries.
Ready to connect★ 335
- View details
knowledge-locatorSkillDocs & knowledge
Lets your agent search and cross-reference information stored in memory palaces using keyword, spatial, and date-based queries.
Ready to connect★ 335
- View details
latent-space-engineeringSkillAI & models
Lets your agent shape its behavior by framing instructions and transferring writing style when composing prompts.
Ready to connect★ 335
- View details
listenhub-cliSkillDocs & knowledge
Routes your agent's requests to the right audio or video skill for tasks like making podcasts, read-aloud, slides, images, or music.
Ready to connect★ 335
- View details
listenhub-voiceSkillMedia
Lets your agent turn text or images into audio, including sound effects, multi-voice dialogue, and cloned voices.
Ready to connect★ 335
- View details
loop-optimizationSkillAI & models
Lets your agent decide when hand-written loop optimizations beat compiler auto-optimizations for hot code.
Ready to connect★ 335
- View details
Make DogfoodSkillAI & models
Lets your agent analyze Makefiles and auto-generate missing build targets.
★ 335
- View details
makefile-generationSkillAI & models
Lets your agent create a Makefile with standard test, lint, format, and build targets for your project.
Ready to connect★ 335
- View details
markdown-formattingSkillFiles & storage
Lets your agent format markdown files with consistent line-wrap and structure for clean git diffs.
Ready to connect★ 335
- View details
mcp-code-executionSkillDatabases & data
Helps your agent run data-heavy, multi-step tool workflows faster by reducing overhead on large datasets.
Ready to connect★ 335
- View details
memory-clarity-probeSkillDocs & knowledge
Lets your agent check whether a session memory or summary is clear enough to hand off or compress.
Ready to connect★ 335
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.
54,624 of the 55,055 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.