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 833 of 1,848
- View details
mock-quality-reviewSkillDev tools
Lets your agent review the quality of test mocks, checking whether they match real API contracts and flagging drift.
Ready to connect★ 221
- View details
model-based-testingSkillMedia
Lets your agent derive test-path candidates from behavior, state, or process models for test design.
Ready to connect★ 221
- View details
multi-agent-testingSkillAI & models
Lets your agent review and improve plans for coordinating multiple AI agents, covering delegation, shared state, conflicts, and termination.
Ready to connect★ 221
- View details
mutation-testing-analysisSkillDev tools
Lets your agent interpret mutation testing reports, explaining which mutants were killed or survived and what evidence exists.
Ready to connect★ 221
- View details
native-mcpSkillAI & models
Built-in MCP (Model Context Protocol) client that connects to external
Ready to connect★ 221
- View details
negative-scenario-discoverySkillDev tools
Lets your agent analyze product designs and evidence to list failure, denial, and degraded-recovery scenarios as NS-## entries.
Ready to connect★ 221
- View details
obliteratusSkillAI & models
Lets your agent modify a language model to remove its refusal behavior using a diff-in-means technique.
Ready to connect★ 221
- View details
observability-design-reviewSkillMonitoring & ops
Lets your agent review logging, metrics, tracing, and alerting designs for gaps before you build them.
Ready to connect★ 221
- View details
ocr-and-documentsSkillDocs & knowledge
Extract text from PDFs and scanned documents. Use web_extract for remote
Ready to connect★ 221
- View details
p5jsSkillDev tools
Lets your agent write p5.js sketches for generative art, shaders, interactive graphics, and 3D.
Ready to connect★ 221
- View details
pairwise-testingSkillMedia
Lets your agent design pairwise test combinations from factors, values, and constraints you provide.
Ready to connect★ 221
- View details
peftSkillAI & models
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods.
Ready to connect★ 221
- View details
performance-bottleneck-analysisSkillDev tools
Helps your agent form evidence-based hypotheses about performance bottlenecks and plan validation steps.
Ready to connect★ 221
- View details
performance-regression-analysisSkillDev tools
Lets your agent compare performance evidence across versions and assess regression risk.
Ready to connect★ 221
- View details
performance-result-analysisSkillDev tools
Lets your agent interpret performance results and evidence quality without inventing conclusions.
Ready to connect★ 221
- View details
performance-test-gatlingSkillDev tools
Lets your agent write and run Gatling performance test simulations for load testing.
Ready to connect★ 221
- View details
performance-test-jmeterSkillMedia
Lets your agent design and run JMeter performance test plans with reports.
Ready to connect★ 221
- View details
performance-test-k6SkillDev tools
Helps your agent write and run k6 load, stress, spike, and soak performance tests.
Ready to connect★ 221
- View details
performance-workload-modelingSkillAI & models
Lets your agent model realistic performance workloads, traffic levels, and acceptance assumptions.
Ready to connect★ 221
- View details
pokemon-playerSkillWeb & browsing
Lets your agent play Pokemon games by controlling a headless emulator and reading game memory.
Ready to connect★ 221
- View details
popular-web-designsSkillWeb & browsing
Gives your agent access to 54 real website design systems as HTML/CSS for reference.
Ready to connect★ 221
- View details
pr-test-impact-analysisSkillDev tools
Lets your agent figure out which tests are affected by a pull request or code diff.
Ready to connect★ 221
- View details
production-incident-analysisSkillMonitoring & ops
Lets your agent analyze production incident evidence, impact, and follow-up actions.
Ready to connect★ 221
- View details
production-verificationSkillDev tools
Helps your agent plan and assess evidence-based verification that a release is working in production.
Ready to connect★ 221
- View details
prompt-injection-testingSkillMedia
Lets your agent design safe prompt-injection tests for AI systems and tool boundaries.
Ready to connect★ 221
- View details
prompt-testingSkillAI & models
Lets your agent test prompt behavior and output changes across prompt versions.
Ready to connect★ 221
- View details
pytorch-fsdpSkillDev tools
Lets your agent train large PyTorch models across multiple GPUs by splitting model weights and data.
Ready to connect★ 221
- View details
quality-dashboard-designSkillMonitoring & ops
Lets your agent design quality dashboards, covering audiences, panels, drill-downs, freshness, and alert rules.
Ready to connect★ 221
- View details
quality-debt-analysisSkillDev tools
Lets your agent analyze quality debt in code or projects, ranking issues by origin, impact, age, priority, and paydown tradeoffs.
Ready to connect★ 221
- View details
quality-gate-designSkillMedia
Lets your agent draft entry criteria, evidence requirements, owners, and exception paths for a release quality gate.
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.
55,004 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.