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
53,895 results · page 130 of 1,797
- View details
redteam-opsecSkillSecurity
Lets your agent follow operational security practices like IP ban evasion, traffic obfuscation, and minimal traces.
Ready to connect★ 7k
- View details
source-code-huntingSkillSecurity
Lets your agent hunt for leaked source code, exposed secrets, and supply-chain issues in codebases.
Ready to connect★ 7k
- View details
specialized-attack-playbooksSkillFiles & storage
Loads step-by-step offensive security playbooks covering CDN misconfig, MITM, and cloud storage attacks.
Ready to connect★ 7k
- View details
unlimited-attack-scopeSkillSecurity
Lets your agent follow a no-limits rule for security testing across web, APK, cloud, AI, wireless, and social engineering.
Ready to connect★ 7k
- View details
web-attack-methodsSkillSecurity
Gives your agent a reference of web attack techniques like SQL injection, XSS, and auth bypass for security testing.
Ready to connect★ 7k
- View details
wireless-hardware-attackSkillSecurity
Lets your agent run wireless and hardware attack techniques like WiFi cracking, Evil Twin, BLE, NFC, SDR, and UART/JTAG exploits.
Ready to connect★ 7k
- View details
zero-day-discoverySkillSecurity
Guides your agent through finding unknown software vulnerabilities using variant analysis, patch diffing, fuzzing, and exploit reasoning.
Ready to connect★ 7k
- View details
Dev BrowserSkillWeb & browsing
Lets your agent open and browse web pages to read their content.
Ready to connect★ 7k
- View details
liger-autopatchSkillAI & models
Lets your agent add Liger Kernel support for a HuggingFace model by generating patch code, tests, and docs.
Ready to connect★ 7k
- View details
liger-kernel-devSkillFiles & storage
Lets your agent write and improve Triton kernels for Liger Kernel from PyTorch code or plain descriptions.
Ready to connect★ 7k
- View details
liger-kernel-perfSkillDocs & knowledge
Lets your agent profile and optimize Liger Kernel GPU code, testing variants and applying the fastest correct one.
Ready to connect★ 7k
- View details
post-exploitationSkillCommerce & finance
Gives your agent offensive security techniques like privilege escalation, lateral movement, and password cracking.
Ready to connect★ 7k
- View details
preline-mcpSkillFiles & storage
Lets your agent add ready-made Preline UI components and page sections into HTML files.
Ready to connect★ 6k
- View details
preline-theme-generatorSkillFiles & storage
Lets your agent generate and check Preline theme CSS files, including light and dark versions matched to your brand colors.
Ready to connect★ 6k
- View details
add-providerSkillAI & models
Lets your agent add a new LLM provider to DeepChat by making reviewed code changes.
Ready to connect★ 6k
- View details
cua-driverSkillFiles & storage
Lets your agent control a macOS app by clicking, typing, and scrolling in its interface.
Ready to connect★ 6k
- View details
deepchat-cliSkillMedia
Lets your agent run DeepChat's command-line tool for text, image, video, and speech generation, transcription, and OCR.
Ready to connect★ 6k
- View details
deepchat-data-importSkillDatabases & data
Helps your agent import, inspect, migrate, or analyze DeepChat chat data and settings.
Ready to connect★ 6k
- View details
deepchat-releaseSkillDocs & knowledge
Lets your agent prepare and publish DeepChat app releases, bumping versions and updating bilingual release notes.
Ready to connect★ 6k
- View details
deepchat-sddSkillDocs & knowledge
Guides your agent to write a durable RFC and execution plan before making substantial DeepChat code or docs changes.
Ready to connect★ 6k
- View details
deepchat-sdd-cleanupSkillDocs & knowledge
Lets your agent tidy project docs by scanning feature, issue, and architecture folders and removing completed issue files.
Ready to connect★ 6k
- View details
deepchat-settingsSkillDev tools
Lets your agent change DeepChat app settings like theme, language, and font size when you ask.
Ready to connect★ 6k
- View details
feishu-toolsSkillFiles & storage
Lets your agent work with Feishu/Lark documents, spreadsheets, and knowledge base content in your workspace.
Ready to connect★ 6k
- View details
git-commitSkillCommunication
Lets your agent write clean, standardized git commit messages for your code changes.
Ready to connect★ 6k
- View details
infographic-syntax-creatorSkillDev tools
Lets your agent turn content into AntV Infographic syntax for rendering infographics.
Ready to connect★ 6k
- View details
memory-managementSkillDocs & knowledge
Guides your agent to remember lessons across chats and route them into memory, skills, scheduled tasks, or tape.
Ready to connect★ 6k
- View details
accessibility-frontend-reviewSkillDev tools
Lets your agent review code changes for accessibility issues like ARIA roles, focus order, and screen reader support.
Ready to connect★ 6k
- View details
androidSkillDev tools
Guides your agent to build the Android version of this project using the right gradle wrapper commands.
Ready to connect★ 6k
- View details
gecko-web-standard-implementationSkillDev tools
Guides your agent to write C++, C, and Rust code that implements web standard interfaces correctly.
Ready to connect★ 6k
- View details
jj-splitSkillDev tools
Guides your agent through splitting one jj commit into two or more separate commits.
Ready to connect★ 6k
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.
53,462 of the 53,895 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.