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
55,436 results · page 832 of 1,848
- View details
circuit-breaker-testingSkillDev tools
Lets your agent plan and review circuit-breaker tests by analyzing open, closed, and half-open states from evidence.
Ready to connect★ 221
- View details
claude-codeSkillProductivity
Lets your agent delegate coding tasks like building features and creating pull requests to the Claude Code CLI.
Ready to connect★ 221
- View details
combinatorial-testingSkillMedia
Lets your agent pick high-risk multi-factor test combinations once factors, values, and constraints are defined.
Ready to connect★ 221
- View details
creative-ideationSkillDev tools
Lets your agent generate ideas using named methods from creative practice.
Ready to connect★ 221
- View details
database-design-quality-reviewSkillDatabases & data
Lets your agent review database schemas and migration plans for design flaws before you build anything.
Ready to connect★ 221
- View details
decision-table-testingSkillMedia
Lets your agent turn business rules and conditions into an auditable table of test-design rule combinations.
Ready to connect★ 221
- View details
dependency-failure-testingSkillDev tools
Guides your agent through analyzing how software dependencies could fail and planning fallback and recovery checks.
Ready to connect★ 221
- View details
distributed-trace-analysisSkillDev tools
Lets your agent analyze distributed traces to find slow calls, errors, and missing steps in request paths.
Ready to connect★ 221
- View details
edge-case-discoverySkillDev tools
Lets your agent find edge cases like boundary values, rare states, and risky combinations from your product docs and test evidence.
Ready to connect★ 221
- View details
equivalence-partitioningSkillMedia
Lets your agent split test inputs into valid, invalid, and unknown classes to design equivalence partitioning test cases.
Ready to connect★ 221
- View details
error-handling-design-reviewSkillMedia
Lets your agent review error handling designs—retries, timeouts, fallbacks, recovery—before code is written.
Ready to connect★ 221
- View details
failover-testingSkillDev tools
Helps your agent plan and review failover tests for primary-secondary systems, covering triggers, traffic transfer, and recovery evidence.
Ready to connect★ 221
- View details
findmySkillDev tools
Lets your agent look up the location of your Apple devices and AirTags using the FindMy app on macOS.
Ready to connect★ 221
- View details
flaky-test-analysisSkillDev tools
Lets your agent investigate intermittent test failures using run history and evidence.
Ready to connect★ 221
- View details
focus-coachSkillDev tools
Lets your agent coach you on choosing a main focus, weighing trade-offs between goals, projects, or learning paths.
Ready to connect★ 221
- View details
full-code-reviewSkillAI & models
Lets your agent run a multi-phase code review of your changes, from design to implementation, before you open a PR.
Ready to connect★ 221
- View details
functional-testingSkillMedia
Lets your agent design functional test plans and test cases for business flows, UI, data, and integrations.
Ready to connect★ 221
- View details
- View details
godmodeSkillAI & models
Lets your agent try prompts that bypass AI model safety filters.
Ready to connect★ 221
- View details
habitSkillDocs & knowledge
Lets your agent create habits and record daily check-ins in your SiYuan task notes.
Ready to connect★ 221
- View details
- View details
kheish-daemon-live-validationSkillDev tools
Return the deterministic readiness marker used by live daemon validation scenarios.
Ready to connect★ 221
- View details
learning-coachSkillAI & models
Lets your agent act as a Chinese-language personal learning coach that tracks lessons and progress across skills like programming and languages.
Ready to connect★ 221
- View details
llm-consistency-testingSkillAI & models
Guides your agent to plan and review LLM consistency tests covering repeat inputs, model versions, prompts, and variance evidence.
Ready to connect★ 221
- View details
llm-evaluation-designSkillDatabases & data
Helps your agent design LLM evaluation datasets, judge criteria, metrics, and human-review boundaries.
Ready to connect★ 221
- View details
llm-hallucination-testingSkillAI & models
Guides your agent to check its answers against evidence, flag unsupported claims, and state uncertainty instead of guessing.
Ready to connect★ 221
- View details
manual-testingSkillDev tools
Helps your agent plan manual and exploratory testing using charters, heuristics, and session records.
Ready to connect★ 221
- View details
metamorphic-testingSkillMedia
Lets your agent design metamorphic test cases by deriving input transformations and expected relations when no direct oracle exists.
Ready to connect★ 221
- View details
metrics-anomaly-analysisSkillMonitoring & ops
Lets your agent detect and investigate unusual spikes or dips in system metrics using observability data.
Ready to connect★ 221
- View details
minecraft-modpack-serverSkillDev tools
Lets your agent set up and host modded Minecraft servers using modpacks from CurseForge or Modrinth.
Ready to connect★ 221
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.
55,005 of the 55,436 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.