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 562 of 1,823
- View details
use-avibe-harnessSkillProductivity
Lets your agent delegate long-running work, schedule tasks, and keep sessions running beyond the current turn.
Ready to connect★ 505
- View details
use-avibe-vaultSkillSecurity
Lets your agent use API keys and passwords to make authenticated requests without ever seeing the secret values.
Ready to connect★ 505
- View details
use-show-pagesSkillDev tools
Lets your agent build, edit, restore, and share visual pages with diagrams, reports, and interactive prototypes.
Ready to connect★ 505
- View details
agdrSkillSearch
Lets your agent search and show AgDR records to check if a similar decision was made before.
Ready to connect★ 502
- View details
analytics-auditSkillDatabases & data
Lets your agent audit your analytics setup, from SDK config to event naming and dashboards.
Ready to connect★ 502
- View details
approve-architectureSkillMedia
Lets your agent record an architect's approval on a pull request to satisfy the architecture review gate.
Ready to connect★ 502
- View details
approve-designSkillMedia
Lets your agent record a designer's approval on a pull request as a gate for merging UI changes.
Ready to connect★ 502
- View details
approve-mergeSkillDev tools
Lets your agent record an explicit per-PR approval and merge that pull request in one step.
Ready to connect★ 502
- View details
audit-depsSkillSecurity
Lets your agent check project dependencies for security vulnerabilities, outdated packages, and license issues.
Ready to connect★ 502
- View details
c4SkillDev tools
Lets your agent generate C4 context and container diagrams in Mermaid from a project's codebase.
Ready to connect★ 502
- View details
codify-ruleSkillDev tools
Turns a code review comment that caught a missed issue into a draft handbook rule entry.
Ready to connect★ 502
- View details
compliance-checkSkillDev tools
Lets your agent audit a website's GDPR and ePrivacy compliance, checking consent, privacy policy, and data handling.
Ready to connect★ 502
- View details
design-syncSkillMedia
Lets your agent sync a local component library to a claude.ai design-system project one component at a time.
Ready to connect★ 502
- View details
dfdSkillSecurity
Lets your agent create data flow diagrams showing trust boundaries and data classifications for threat modeling.
Ready to connect★ 502
- View details
docs-auditSkillCloud & infra
Lets your agent audit documentation for coverage of tutorials, how-tos, references, explanations, and staleness.
Ready to connect★ 502
- View details
element-backwards-compatSkillDev tools
Gives your agent the rules for keeping PrairieLearn elements backwards compatible when their controllers change.
Ready to connect★ 502
- View details
element-runtime-assetsSkillDev tools
Helps your agent create PrairieLearn elements or update their scripts, styles, and bundled assets.
Ready to connect★ 502
- View details
element-validationSkillDev tools
Lets your agent validate and review PrairieLearn element question.html contracts, schemas, and Python checks.
Ready to connect★ 502
- View details
eval-agentsSkillMonitoring & ops
Lets your agent score a code-review agent's accuracy against a labeled set of pull requests using ground-truth defects.
Ready to connect★ 502
- View details
express-request-validationSkillDev tools
Gives your agent conventions for validating Express request inputs with Zod in PrairieLearn.
Ready to connect★ 502
- View details
extract-featuresSkillAI & models
Lets your agent inventory an app's routes, models, jobs, tests, UI, and docs into a spec for rewrites.
Ready to connect★ 502
- View details
fan-outSkillCommunication
Lets your agent run many parallel sub-agent tasks at once, each isolated in its own workspace.
Ready to connect★ 502
- View details
feature-diagramSkillAI & models
Lets your agent draw a Mermaid flowchart for each feature, showing its routes, models, jobs, or screens.
Ready to connect★ 502
- View details
geo-auditSkillWeb & browsing
Lets your agent check how well a website is set up for AI search engines and web crawlers.
Ready to connect★ 502
- View details
groups-and-teamsSkillDev tools
Helps your agent work with groups and teams in PrairieLearn and understand its table and variable names.
Ready to connect★ 502
- View details
handoverSkillDev tools
Lets your agent assess an inherited codebase and produce a structured handover document with a harnessability score.
Ready to connect★ 502
- View details
ideaSkillDev tools
Lets your agent save new product ideas, feature concepts, or tool proposals to an ideas backlog.
Ready to connect★ 502
- View details
inboxSkillDev tools
Lets your agent show all items across your projects that need your attention, like PRs, issues, comments, and blockers.
Ready to connect★ 502
- View details
launch-checkSkillDatabases & data
Lets your agent run a pre-launch readiness check across ten areas like security, performance, SEO, and monitoring.
Ready to connect★ 502
- View details
monitoring-auditSkillMonitoring & ops
Lets your agent audit a project's monitoring setup, including logging, error tracking, health checks, alerting, and runbooks.
Ready to connect★ 502
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.