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.
Category: AI & models
9,032 results · page 236 of 302
- View details
oracle-grill-meSkillAI & models
Grill the user relentlessly about a plan, decision, or idea, maintaining the project's domain model (CONTEXT.md, ADRs) as decisions crystallise. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.
Ready to connect
- View details
p0a-micro-needs-detectorSkillAI & models
Micro-need five-questions detector. Teams naturally favor "big requirements"; this Skill brings small-looking but potentially deep problems back into view. Based on《AI rebuild product
Ready to connect
- View details
p0b-real-needs-validatorSkillAI & models
A five-question validator for judging real needs. In the AI era, the greatest danger is not the absence of needs, but that fake needs so easily masquerade as real ones. Based on the 'AI rebuild product needs' tool card.
Ready to connect
- View details
p0c-needs-decomposerSkillAI & models
Four-layer needs decomposition card. Many teams, upon hearing user input, immediately break it down into features. This card forces the team to first separate "what the user said" from "what the user is actually stuck on." Based on "AI rebuild
Ready to connect
- View details
p0d-needs-archaeologistSkillAI & models
The five-step requirements archaeology method. Requirements as stated by users are just the tip of the iceberg; real requirements are often buried in historical decisions, organizational structures, and business evolution. This Skill teaches you how to dig up deep requirements.
Ready to connect
- View details
p0e-good-question-generatorSkillAI & models
Six-dimension observation table for good questions. In the AI era, a product manager's most important skill is not giving answers but asking good questions. This Skill provides a six-dimension observation framework to help you discover good questions from different angles.
Ready to connect
- View details
p0f-agent-boundary-designerSkillAI & models
Agent boundary checklist. The biggest risk with AI products isn't underperformance—it's overreach. This Skill helps you systematically define what your AI should do, what it shouldn't do, and when it must stop and wait for a human.
Ready to connect
- View details
p0g-diverse-recommendation-rewriterSkillAI & models
Diverse recommendation rewriting checklist. In the AI era, recommendation systems are no longer about "guessing what you like", but "helping you discover what you didn't know you'd like". This Skill provides a framework for rewriting single recommendations into diverse recommendations.
Ready to connect
- View details
p0g-diversity-rewrite-checklistSkillAI & models
A checklist for diverse recommendation rewrites. When teams rewrite their recommendation system, the easiest path is just to add a bit of randomness on top of the original objective function. This Skill helps you avoid 'looking more diverse on the surface while remaining monolithic underneath'.
Ready to connect
- View details
p0h-triple-balance-assessorSkillAI & models
Triple balance sheet for AI products. When teams discuss AI product direction, they most often focus only on growth and efficiency. This Skill helps teams look at all three layers at once: business, humanity, and culture. Based on 'AI
Ready to connect
- View details
- View details
- View details
- View details
- View details
- View details
p12-contemplation-orchestratorSkillAI & models
Orchestrator for the full 'Contemplation' book — routes to the corresponding chapter Skill based on the type of question entered by the user, chaining together the full end-to-end decision-correction workflow
Ready to connect
- View details
p12a-contemplation-prerequisite-checkSkillAI & models
Checks whether the prerequisites for using a method hold — identifies situational changes, verifies assumed premises, and reconfigures the applicable method.
Ready to connect
- View details
p12a-contemplation-right-actionSkillAI & models
First check whether the mechanism is harmful, then whether the outcome is effective — a fourfold check of value/cost/emotion/right to exit
Ready to connect
- View details
p12a-contemplation-right-effortSkillAI & models
Stopping a wrong direction is also progress—reset-to-zero analysis, pause strategies, and stop-loss decisions
Ready to connect
- View details
p12a-contemplation-right-livelihoodSkillAI & models
Revenue sources reveal where a product's loyalty lies — a review of income sources and incentive bias check.
Ready to connect
- View details
p12a-contemplation-right-mindfulnessSkillAI & models
Continuous awareness, systematic reflection, and consolidating judgments — building decision-making awareness for individuals and teams
Ready to connect
- View details
p12a-contemplation-right-thinkingSkillAI & models
Fix the judgment path instead of piling on more opinions — break down the chain of judgment and distinguish premises, evidence, reasoning, and emotions.
Ready to connect
- View details
p12a-contemplation-right-viewSkillAI & models
See the problem accurately, not just explain away the symptoms — the three-layer method for viewing problems (phenomenon layer / context layer / relationship layer)
Ready to connect
- View details
p12a-contemplation-view-correctionSkillAI & models
Correct the viewpoint first, then invoke the method — default checks, evidence validation, consequence questioning, and eight correction perspectives
Ready to connect
- View details
p13-intuition-orchestratorSkillAI & models
Orchestrator for the entire book 'Judgment and Intuition' — a router and aggregator for the nine-step closed-loop decision roadmap
Ready to connect
- View details
p13a-judgment-metacognitionSkillAI & models
Metacognition of judgment — understanding what judgment is and recognizing the scenarios and types of judgment
Ready to connect
- View details
p13b-systemic-thinkerSkillAI & models
Systemic thinking — structured analysis, relationship mapping, feedback loop identification
Ready to connect
- View details
p13d-intuition-trainingSkillAI & models
Intuition training — transforming intuition into compressible cognitive models and pattern recognition
Ready to connect
- View details
p13e-nine-step-frameworkSkillAI & models
Overview of the nine-step closed loop for AI product decisions—covering what to build, whether it's worth it, whether to use AI, what to achieve, user trust, how to do it, how to validate, how to sell, and the feedback loop
Ready to connect
- View details
p13f-first-half-judgmentSkillAI & models
First half of nine steps — a judgment framework for what to do, whether it's worth it, and whether to use AI
Ready to connect
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.
54,855 of the 55,286 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.