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
54,017 results · page 90 of 1,801
- View details
storage-formatSkillDatabases & data
Teaches your agent how database files are organized on disk, from pages to B-trees to freelist.
Ready to connect★ 24k
- View details
transaction-correctnessSkillDatabases & data
Teaches your agent how Turso's database handles transactions, WAL logging, checkpoints, and crash recovery.
Ready to connect★ 24k
- View details
yield-injectionsSkillDatabases & data
Lets your agent add yield points and simulated failures to state machines so long tasks can pause and resume safely.
Ready to connect★ 24k
- View details
furniture-fitSkillDev tools
Lets your agent check whether furniture like sofas or beds will fit in a room based on measurements.
Ready to connect★ 24k
- View details
open-prSkillDev tools
Lets your agent open a pull request in the pascalorg/editor repo using that repo's PR template.
Ready to connect★ 24k
- View details
open-pr2SkillDev tools
Lets your agent open or update a pull request on pascalorg/editor with a plain-language issue-and-fix description from the branch diff.
Ready to connect★ 24k
- View details
pascal-3dSkillAI & models
Lets your agent create, edit, inspect, and validate 3D building and room models in Pascal.
Ready to connect★ 24k
- View details
review-architectureSkillAI & models
Lets your agent review code changes against the Pascal project's architecture rules like package boundaries and hook usage.
Ready to connect★ 24k
- View details
control-flowSkillMedia
Lets your agent read source code and draw ASCII tree diagrams of control flows and data structures.
Ready to connect★ 24k
- View details
devSkillDev tools
Lets your agent build, install, test, and run the Happy app locally using step-by-step dev instructions.
Ready to connect★ 24k
- View details
maintainSkillDev tools
Lets your agent triage GitHub issues, find duplicates, and draft replies for the slopus/happy project with your approval.
Ready to connect★ 24k
- View details
office-hoursSkillMedia
Runs a structured brainstorming session that questions your product idea and saves a design doc.
Ready to connect★ 24k
- View details
sessionsSkillSearch
Lets your agent search and answer questions about past coding agent session history across Claude Code, Codex, and Cursor.
Ready to connect★ 24k
- View details
terminal-emulatorSkillMedia
Lets your agent test interactive CLI and terminal apps by sending keystrokes and checking screen output.
Ready to connect★ 24k
- View details
nc-hermes-e2eSkillDev tools
Lets your agent verify that the Hermes skill is discoverable and works in a fresh session.
Ready to connect★ 22k
- View details
nemoclaw-contributor-onboardSkillDev tools
Lets your agent set up and verify a NemoClaw development checkout so you can contribute code.
Ready to connect★ 22k
- View details
nemoclaw-contributor-plan-issueSkillDev tools
Lets your agent break a GitHub issue into smaller, independently valuable pieces before coding starts.
Ready to connect★ 22k
- View details
nemoclaw-contributor-update-dependenciesSkillSecurity
Lets your agent audit and apply a dependency upgrade, tracing upstream changes into your codebase and resolving security concerns.
Ready to connect★ 22k
- View details
nemoclaw-contributor-update-docsSkillDocs & knowledge
Lets your agent find merged user-visible changes and update the matching documentation pages.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-analyze-ci-performanceSkillMedia
Lets your agent analyze CI timing data to find slow CLI tests or slow base-image publication steps.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-analyze-pr-value-streamSkillMonitoring & ops
Lets your agent analyze a pull request's full timeline and find where time is spent and bottlenecks occur.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-classify-ci-failureSkillMonitoring & ops
Lets your agent classify a failing GitHub Actions CI job from redacted logs.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-cross-issue-sweepSkillFiles & storage
Lets your agent scan open issues to find which ones a pull request also fixes or conflicts with, with file evidence.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-cut-release-tagSkillDev tools
Lets your agent cut a signed semver release tag after candidate checks and draft the release announcement.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-daySkillSecurity
Lets your agent run a scheduled maintainer pass that picks workflows to merge PRs, fix security issues, and close them.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-e2eSkillDev tools
Lets your agent start trusted GitHub Actions end-to-end test runs and report their results.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-eveningSkillDev tools
Lets your agent run the end-of-day release handoff and optionally cut a release tag.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-find-review-prSkillSecurity
Lets your agent find urgent, security-labeled PRs to review and spot duplicates.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-fix-e2e-failuresSkillMonitoring & ops
Lets your agent triage and fix failing E2E test runs by grouping root causes, opening and merging PRs, and watching new results.
Ready to connect★ 22k
- View details
nemoclaw-maintainer-morningSkillDev tools
Runs a morning standup: triages the backlog, picks the day's target version, labels items, and outputs a daily plan.
Ready to connect★ 22k
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.
53,585 of the 54,017 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.