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 169 of 1,801
- View details
kiln-check-deprecationSkillAI & models
Lets your agent check a model list for deprecated or end-of-life AI models across providers.
Ready to connect★ 5k
- View details
kiln-check-finetune-deprecationSkillAI & models
Lets your agent check if your fine-tuning models are deprecated or no longer supported.
Ready to connect★ 5k
- View details
kiln-prerelease-checkSkillDev tools
Lets your agent run prerelease smoke tests and CI checks on a release candidate and write a report of failures with fixes.
Ready to connect★ 5k
- View details
playwrightSkillWeb & browsing
Lets your agent open the Kiln app in a real browser, take screenshots, and run its end-to-end tests.
Ready to connect★ 5k
- View details
release-digestSkillCommunication
Lets your agent post a Slack recap of changes since the last release, grouping merged PRs by author and type.
Ready to connect★ 5k
- View details
ticket-routing-playbookSkillDev tools
Lets your agent assign an owning team and priority level to incoming support tickets.
Ready to connect★ 5k
- View details
fde-ontologySkillCloud & infra
Lets your agent work with industrial IoT device data over Modbus, OPC UA, and MQTT.
Ready to connect★ 5k
- View details
macroSkillDev tools
Guides your agent to follow the right approach when writing procedural macros for the Topcoat Rust library.
Ready to connect★ 5k
- View details
proseSkillDocs & knowledge
Guides your agent to follow Topcoat's style rules when writing long markdown documentation.
Ready to connect★ 5k
- View details
styleSkillDocs & knowledge
Gives your agent the house style rules for writing Rust code and docs in the Topcoat repository.
Ready to connect★ 5k
- View details
abstractionSkillMedia
Lets your agent distill a complex idea or workflow into the simplest complete form needed for a decision.
Ready to connect★ 5k
- View details
agent-goalSkillAI & models
Lets your agent write a /goal line that sets an objective, validation evidence, and a stop condition for long-running coding work.
Ready to connect★ 5k
- View details
agent-instructionsSkillMedia
Lets your agent write, check, and clean up instruction files like AGENTS.md and CLAUDE.md.
Ready to connect★ 5k
- View details
arktypeSkillDev tools
Helps your agent write Arktype code for runtime type validation with unions and schema merging.
Ready to connect★ 5k
- View details
asymmetric-winsSkillMonitoring & ops
Lets your agent spot when a small feature request would add outsized complexity and argue to decline it.
Ready to connect★ 5k
- View details
authSkillSecurity
Guides your agent when it edits Epicenter's auth code, handling OAuth sessions, identity state, and hosted sign-in.
Ready to connect★ 5k
- View details
autumnSkillDev tools
Lets your agent manage billing tasks like checking credits, gating plans, issuing refunds, and tracking AI usage.
Ready to connect★ 5k
- View details
better-auth-security-best-practicesSkillCloud & infra
Lets your agent review and harden authentication security like rate limits, secrets, CSRF, cookies, and session handling.
Ready to connect★ 5k
- View details
change-proposalSkillFiles & storage
Lets your agent present a proposed code change as side-by-side diffs, diagrams, or comparison tables.
Ready to connect★ 5k
- View details
cloudflare-workersSkillCloud & infra
Gives your agent knowledge of Cloudflare Workers patterns for building and deploying serverless edge code.
Ready to connect★ 5k
- View details
code-auditSkillDev tools
Finds recurring code smells in Epicenter code and scopes the cleanup they need.
Ready to connect★ 5k
- View details
code-proseSkillDev tools
Lets your agent rewrite code identifiers and comments following Orwell's style rules without changing behavior.
Ready to connect★ 5k
- View details
codex-task-backlog-hygieneSkillProductivity
Lets your agent audit Codex tasks, save unowned outcomes to BACKLOG.md, and close finished chats.
Ready to connect★ 5k
- View details
cohesion-over-testabilitySkillDev tools
Refactors your code to remove abstractions added only for testing, keeping behavior and coverage intact.
Ready to connect★ 5k
- View details
collapse-passSkillDev tools
Lets your agent simplify code by removing unnecessary abstraction layers across a diff or package.
Ready to connect★ 5k
- View details
consult-claudeSkillAI & models
Lets your agent get a second opinion from Claude to review code or research a topic independently.
Ready to connect★ 5k
- View details
define-errorsSkillCommunication
Gives your agent patterns and helpers for defining error types and extracting error messages in code.
Ready to connect★ 5k
- View details
dialecticSkillAI & models
Lets your agent work through a fuzzy idea by voicing and testing opposing concrete explanations until the model becomes clear.
Ready to connect★ 5k
- View details
drizzle-ormSkillDatabases & data
Lets your agent write database code using Drizzle ORM patterns for schemas, migrations, and typed queries.
Ready to connect★ 5k
- View details
factory-function-compositionSkillDev tools
Teaches your agent factory function patterns for composing clients and services in code.
Ready to connect★ 5k
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.