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 907 of 1,848
- View details
trace-mcp-pre-commitSkillSecurity
Lets your agent run security, quality, and antipattern checks on code before committing or opening a pull request.
Ready to connect★ 178
- View details
trace-mcp-refactoringSkillFiles & storage
Lets your agent rename code symbols across all files while checking impact and finding safe refactor candidates.
Ready to connect★ 178
- View details
answersSkillSearch
Lets your agent answer questions with web search results and citations.
Ready to connect★ 177
- View details
automating-the-browserSkillWeb & browsing
Lets your agent interact with web pages by clicking, typing, scrolling, and running scripts before extracting the page content.
Ready to connect★ 177
- View details
build-for-testingSkillFiles & storage
Lets your agent compile the TMDb package and all test targets to catch test-code build errors without running the tests.
Ready to connect★ 176
- View details
canon-tddSkillAI & models
Guides your agent to write code test-first: list tests, write one failing test, make it pass, refactor, repeat.
Ready to connect★ 176
- View details
clean-skillSkillSecurity
A harmless test skill that does nothing, used to verify security scanners report no false findings.
Ready to connect★ 177
- View details
crawlbergSkillWeb & browsing
Lets your agent crawl, scrape, and convert web pages to Markdown, following links or driving a real browser.
Ready to connect★ 177
- View details
crawling-a-siteSkillWeb & browsing
Lets your agent crawl a website by following links and saving every reachable page as Markdown.
Ready to connect★ 177
- View details
deliverSkillAI & models
Lets your agent turn an approved plan into a finished pull request: implement, test, review, and open it.
Ready to connect★ 176
- View details
dev-cycleSkillWeb & browsing
Lets your agent run and debug crawlberg build, install, format, and e2e test cycles via ready-made Taskfile tasks.
Ready to connect★ 177
- View details
diagnose-ci-failureSkillDocs & knowledge
Lets your agent diagnose a failing CI run and report which job failed, why, and a concrete fix.
Ready to connect★ 176
- View details
diagnose-integration-failureSkillDev tools
Lets your agent diagnose a failing TMDb integration-test run by summarising the failure, ranking likely causes, and suggesting a fix.
Ready to connect★ 176
- View details
document-swiftSkillDocs & knowledge
Lets your agent write Swift DocC documentation comments for public code following the TMDb package's style rules.
Ready to connect★ 176
- View details
fix-integration-failuresSkillProductivity
Lets your agent diagnose and fix a failing scheduled Integration test run, retrying transient failures or patching real issues via a PR.
Ready to connect★ 176
- View details
fix-pr-checksSkillAI & models
Lets your agent diagnose and fix failing CI checks on a pull request, then commit and push the fixes.
Ready to connect★ 176
- View details
forensifySkillFiles & storage
Lets your agent audit all installed AI agent tools, plugins, and credentials on your machine and produce a security briefing.
Ready to connect★ 177
- View details
headless-fallbackSkillWeb & browsing
Lets your agent load JavaScript-heavy web pages with a real browser when plain fetching returns nothing useful.
Ready to connect★ 177
- View details
images-searchSkillSearch
Lets your agent search the web for images and get titles, source links, and thumbnails.
Ready to connect★ 177
- View details
llm-contextSkillSearch
Lets your agent fetch web pages as clean text, tables, and code ready to feed into AI answers.
Ready to connect★ 177
- View details
local-descriptionsSkillSearch
Lets your agent fetch AI-generated descriptions for places using place IDs from search results.
Ready to connect★ 177
- View details
local-place-searchSkillSearch
Lets your agent find local places like businesses, addresses, and cities with ratings, hours, and phone numbers.
Ready to connect★ 177
- View details
local-poisSkillSearch
Lets your agent look up full details of local businesses like ratings, hours, and contact info using place IDs.
Ready to connect★ 177
- View details
malicious-skillSkillDev tools
Claims to add task and calendar productivity features but contains a hidden prompt injection instruction.
Ready to connect★ 177
- View details
mapping-urlsSkillWeb & browsing
Lets your agent list the URLs on a website by analyzing its sitemap without crawling full page content.
Ready to connect★ 177
- View details
news-searchSkillSearch
Lets your agent search recent news articles with filters for date, safety, and custom ranking.
Ready to connect★ 177
- View details
poisoned-skillSkillProductivity
Claims to help with daily tasks but its description contains a prompt injection telling the agent to hide risks.
Ready to connect★ 177
- View details
review-changesSkillAI & models
Lets your agent review code changes on the current branch and return a severity-graded report of issues.
Ready to connect★ 176
- View details
review-knowledgeSkillAI & models
Lets your agent audit your project's knowledge notes and skill files for stale, contradictory, or drifted claims.
Ready to connect★ 176
- View details
review-pr-threadsSkillDev tools
Lets your agent work through unresolved PR review comments, fixing issues or replying, then resolves each thread.
Ready to connect★ 176
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.