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 575 of 1,823
- View details
catwoe-analysisSkillDev tools
Lets your agent apply CATWOE analysis to a problem from a chosen stakeholder's perspective.
Ready to connect★ 481
- View details
causal-chain-querySkillMonitoring & ops
Lets your agent trace how one event causes another by following cause-and-effect links step by step.
Ready to connect★ 481
- View details
causal-chain-tracingSkillMonitoring & ops
Lets your agent trace a problem back to its root cause using IF...THEN...BECAUSE logic chains.
Ready to connect★ 481
- View details
causal-claim-extractionSkillDev tools
Lets your agent find and pull out cause-and-effect claims like 'X causes Y' from text.
Ready to connect★ 481
- View details
causal-modelingSkillAI & models
Lets your agent build causal models by identifying variables and mapping how mechanisms connect.
Ready to connect★ 481
- View details
causal-necessity-testingSkillDev tools
Lets your agent extract causal claims from text and estimate how likely each cause was necessary for the outcome.
Ready to connect★ 481
- View details
causal-tree-buildingSkillDev tools
Lets your agent build causal trees that trace symptoms back to root causes.
Ready to connect★ 481
- View details
challenge-operationSkillSecurity
Lets your agent question current practices by asking non-threatening 'why' questions using the de Bono Challenge method.
Ready to connect★ 481
- View details
challenge-questioningSkillSecurity
Lets your agent ask gentle 'why' questions about current practices to uncover the history behind them.
Ready to connect★ 481
- View details
checkpoint-and-recoverSkillDev tools
Lets your agent save checkpoints before risky operations and recover state if something goes wrong.
Ready to connect★ 481
- View details
circular-validation-auditSkillDev tools
Lets your agent check whether a validation test would only confirm a theory because it assumes the theory's own premises.
Ready to connect★ 481
- View details
citation-chainingSkillMonitoring & ops
Lets your agent trace a paper's citations forward and backward to find more related research.
Ready to connect★ 481
- View details
citation-network-analysisSkillDev tools
Lets your agent build and analyze patent citation networks using methods like main path analysis and PageRank.
Ready to connect★ 481
- View details
claim-analysisSkillDev tools
Lets your agent break down patent claims into independent and dependent parts to analyze their scope.
Ready to connect★ 481
- View details
claim-extractionSkillDev tools
Lets your agent pull factual claims out of source material using a structured extraction strategy.
Ready to connect★ 481
- View details
claim-negationSkillDev tools
Lets your agent formally negate a claim to produce its logical complement.
Ready to connect★ 481
- View details
claim-page-creationSkillDev tools
Walks your agent through creating a claim page in the vault following a step-by-step procedure.
Ready to connect★ 481
- View details
claim-parsingSkillDev tools
Lets your agent parse patent claim syntax and map independent versus dependent claim relationships.
Ready to connect★ 481
- View details
claim-refinementSkillDev tools
Lets your agent refine a claim so it survives counterexamples while staying true to the original idea.
Ready to connect★ 481
- View details
clarify-resourcesSkillDev tools
Helps your agent figure out what research resources like compute and data the user has available.
Ready to connect★ 481
- View details
classification-navigationSkillDev tools
Lets your agent explore patent classification categories, going deeper into a branch or sideways to related ones.
Ready to connect★ 481
- View details
closest-worldsSkillDev tools
Lets your agent find the smallest change to a described situation that would make a goal true.
Ready to connect★ 481
- View details
clr-validationSkillDev tools
Lets your agent check its reasoning for flaws using Goldratt's eight categories of legitimate reservation.
Ready to connect★ 481
- View details
cluster-analysisSkillDev tools
Lets your agent find natural opinion clusters among collected judgments and describe them.
Ready to connect★ 481
- View details
coherence-diagnosisSkillDev tools
Lets your agent audit whether a set of preferences is internally consistent using a consistency ratio.
Ready to connect★ 481
- View details
cold-startSkillDev tools
Gives your agent a strategy to pick a research direction when you have none.
Ready to connect★ 481
- View details
collective-adjudicationSkillDev tools
Lets your agent combine rankings from multiple judges into one fair final ranking using voting methods.
Ready to connect★ 481
- View details
combination-enumerationSkillDev tools
Lets your agent systematically list all combinations of options across multiple dimensions.
Ready to connect★ 481
- View details
combination-evaluationSkillDev tools
Lets your agent assess whether combining ideas or approaches is workable and genuinely new.
Ready to connect★ 481
- View details
combinatorial-creativitySkillDev tools
Lets your agent generate new ideas by combining existing concepts into fresh ones.
Ready to connect★ 481
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.