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 931 of 1,848

  • protenixSkillMedia

    Lets your agent predict 3D structures of protein complexes using an open AlphaFold3-style model.

    Ready to connect★ 162

    github.com/adaptyvbio/protein-design-skills162 stars

    View details
  • realignSkillDev tools

    Lets your agent refocus a project's direction using a three-horizon planning method via /realign.

    Ready to connect★ 162

    github.com/jongwony/epistemic-protocols161 stars

    View details
  • realizeSkillDev tools

    Lets your agent run runtime checks that a protocol's type transitions actually work as specified.

    Ready to connect★ 162

    github.com/jongwony/epistemic-protocols161 stars

    View details
  • recollectSkillDev tools

    Lets your agent recover lost context by scanning past work and confirming the right memories with you.

    Ready to connect★ 162

    github.com/jongwony/epistemic-protocols161 stars

    View details
  • reduced-space-testSkillDev tools

    Lets your agent test whether a cheaper stand-in method matches a target by running bounded, evidence-based checks.

    Ready to connect★ 162

    github.com/jongwony/epistemic-protocols161 stars

    View details
  • skillhone-benchmark-optimizationSkillDev tools

    Lets your agent run benchmark optimization workflows to build eval sets, set baselines, and compare skills.

    Ready to connect★ 162

    github.com/tencent/skillhone161 stars

    View details
  • sublateSkillDev tools

    Lets your agent challenge its own assumptions before acting.

    Ready to connect★ 162

    github.com/jongwony/epistemic-protocols161 stars

    View details
  • white-bearSkillAI & models

    Lets your agent audit text for unnecessary mentions of competing targets and other negative framing.

    Ready to connect★ 162

    github.com/jongwony/epistemic-protocols161 stars

    View details
  • zero-shotSkillAI & models

    Lets your agent audit text for over-reliance on few-shot examples instead of stated principles.

    Ready to connect★ 162

    github.com/jongwony/epistemic-protocols161 stars

    View details
  • abort-safe-handlerSkillProductivity

    Lets your agent skip a question-answering loop cleanly, with no partial state left behind.

    Ready to connect★ 161

    github.com/fr-e-d/gaai-framework147 stars

    View details
  • ai-changelogSkillMonitoring & ops

    Lets your agent set up a CHANGELOG.md file and rules so it keeps a record of changes in any project.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • analyse-designSkillWeb & browsing

    Lets your agent analyse a website or app's design and extract its colours, fonts, spacing and UI patterns.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • apply-mantel-stylesSkillDev tools

    Gives your agent Mantel brand style guidelines so diagrams and frontend components match Mantel's look.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • authoring-claude-mdSkillFiles & storage

    Guides your agent in writing and updating CLAUDE.md files that store project context and coding conventions.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • aws-strands-agents-agentcoreSkillCloud & infra

    Gives your agent expert guidance for building and deploying AI agents with AWS Strands Agents SDK and Amazon Bedrock AgentCore.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • backlogSkillFiles & storage

    Lets your agent capture bug reports and feature requests, or list and triage a project backlog.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • claude-agent-sdkSkillAI & models

    Gives your agent guidance on building apps with the Claude Agent SDK, including patterns and pitfalls.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • code-review-recent-changesSkillDev tools

    Lets your agent review recent code changes for standards, spec compliance, and maintainability with a severity-ordered verdict.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • create-modeSkillDocs & knowledge

    Lets your agent create and scaffold new Pneuma modes with manifests, viewers, skills, and docs.

    Ready to connect★ 161

    github.com/pandazki/pneuma-skills148 stars

    View details
  • creating-development-plansSkillDocs & knowledge

    Lets your agent create structured development plans with phased tasks, requirements, and QA checklists.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • critical-thinking-logical-reasoningSkillDev tools

    Lets your agent critically analyse the reasoning and arguments in written content like articles and reports.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • diagnose-hard-problemSkillDev tools

    Guides your agent through a step-by-step process to find the root cause of stubborn bugs.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • diagrams-as-codeSkillDev tools

    Lets your agent create architecture diagrams as code using Python's diagrams library.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • differed-task-executionSkillProductivity

    Lets your agent delay a task until a set time or after a wait, like running something in 30 minutes.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • excalidraw-diagramsSkillDev tools

    Lets your agent create and update Excalidraw diagrams following best practices.

    Ready to connect★ 161

    github.com/sammcj/agentic-coding159 stars

    View details
  • geoSkillWeb & browsing

    Lets your agent handle search-optimization requests like diagnosing a site, planning content, and checking visibility in AI search results.

    Ready to connect★ 160

    github.com/yaojingang/geohub161 stars

    View details
  • geo-contentSkillDocs & knowledge

    Lets your agent write evidence-based titles, explainers, comparisons, rankings, and page outlines in Markdown.

    Ready to connect★ 160

    github.com/yaojingang/geohub161 stars

    View details
  • geo-diagnoseSkillWeb & browsing

    Lets your agent audit a website or page and find gaps in how well AI search engines can find and cite it.

    Ready to connect★ 160

    github.com/yaojingang/geohub161 stars

    View details
  • geo-discoverSkillSearch

    Lets your agent find search questions, keyword ideas, and content topics from a brief in Chinese or English.

    Ready to connect★ 160

    github.com/yaojingang/geohub161 stars

    View details
  • geo-knowledgeSkillDev tools

    Lets your agent build and query a knowledge graph from your sources, linking facts to their evidence.

    Ready to connect★ 160

    github.com/yaojingang/geohub161 stars

    View details

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.

See how to connect your AI