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
54,221 results · page 389 of 1,808
- View details
time-zone-safetySkillDatabases & data
Flags code that uses current-time calls or compares across timezones so your agent can fix timezone bugs before they cause wrong statistics.
Ready to connect★ 1k
- View details
kotlin-backend-jpa-entity-mappingSkillMedia
Guides your agent to write correct Kotlin entity code for Spring Data JPA and Hibernate.
Ready to connect★ 1k
- View details
astSkillDev tools
Lets your agent build abstract syntax trees manually using FunctionBuilder for kernels and callables.
Ready to connect★ 1k
- View details
backend_architectureSkillDev tools
Gives your agent reference docs on backend plugin architecture, device interfaces, dynamic loading, and command encoding.
Ready to connect★ 1k
- View details
cmakeSkillDev tools
Gives your agent guidance on CMake build options and patterns for building LuisaCompute.
Ready to connect★ 1k
- View details
glslangSkillDev tools
Lets your agent build SPIR-V shader code using GLSL types, instructions, control flow, and decorations.
Ready to connect★ 1k
- View details
hlslSkillAI & models
Lets your agent write HLSL shader code using StringBuilder patterns, builtin headers, and DXIL embedding.
Ready to connect★ 1k
- View details
ir_pipelineSkillDev tools
Lets your agent compile code through legacy IR and XIR pipelines with AST lowering, SSA IR, and optimization passes.
Ready to connect★ 1k
- View details
lc_astSkillDev tools
Lets your agent find AST variable usage markers and CallOp rules in LuisaCompute source code.
Ready to connect★ 1k
- View details
lc_coreSkillMonitoring & ops
Lets your agent use a core code library with logging, math, memory allocators, and container utilities.
Ready to connect★ 1k
- View details
lc_dslSkillDev tools
Lets your agent write and run custom GPU compute kernels for parallel processing tasks.
Ready to connect★ 1k
- View details
lc_optimizeSkillDocs & knowledge
Optimizes LuisaCompute GPU kernels for faster performance by improving atomics, reductions, and thread communication.
Ready to connect★ 1k
- View details
lc_runtimeSkillMonitoring & ops
Lets your agent access low-level graphics and compute APIs for devices, streams, images, and ray tracing.
Ready to connect★ 1k
- View details
lc_vstlSkillDev tools
Gives your agent access to Vyper-style container utilities like HashMaps, queues, pools, variants, and smart pointers.
Ready to connect★ 1k
- View details
llvm-spirvSkillDev tools
Lets your agent translate LLVM intermediate code into SPIR-V for Vulkan GPU execution with validation checks.
Ready to connect★ 1k
- View details
project_structureSkillMedia
Gives your agent reference knowledge about project layout, module architecture, compiler pipeline, and design patterns.
Ready to connect★ 1k
- View details
rust_workspaceSkillDev tools
Gives your agent knowledge of Rust compiler internals, FFI, and crate structure for development tasks.
Ready to connect★ 1k
- View details
spirv_codegenSkillDev tools
Lets your agent convert Vulkan shader code into SPIR-V binaries with validation and legalization.
Ready to connect★ 1k
- View details
spv-optSkillDev tools
Lets your agent develop and test SPIR-V optimizer passes and manipulate shader IR.
Ready to connect★ 1k
- View details
xir_passesSkillDev tools
Lets your agent write and edit XIR transformation passes in src/xir/passes/.
Ready to connect★ 1k
- View details
xmakeSkillDev tools
Gives your agent knowledge of XMake build configuration and commands for the LuisaCompute project.
Ready to connect★ 1k
- View details
yyjsonSkillDev tools
Lets your agent parse, create, and modify JSON data quickly using a C library.
Ready to connect★ 1k
- View details
basic-syntaxSkillMedia
Gives your agent the coding style rules for Trezor Suite so it writes consistent TypeScript and React code.
Ready to connect★ 1k
- View details
commentsSkillDev tools
Gives your agent conventions for writing clear, minimal code comments when working on Trezor Suite code.
Ready to connect★ 1k
- View details
common-issuesSkillDev tools
Lets your agent look up fixes for known build, test, and environment problems in the Trezor Suite codebase.
Ready to connect★ 1k
- View details
common-tasksSkillProductivity
Lets your agent manage dependencies, create packages, and troubleshoot builds in the Trezor Suite codebase.
Ready to connect★ 1k
- View details
defensive-programmingSkillDev tools
Teaches your agent to write safer TypeScript with type checks, explicit return types, and careful error handling.
Ready to connect★ 1k
- View details
dependency-injectionSkillDev tools
Helps your agent write code that follows the dependency injection pattern used in a codebase, with consistent service definitions and wiring.
Ready to connect★ 1k
- View details
development-commandsSkillDev tools
Lets your agent run, build, lint, and test code in the Trezor Suite monorepo.
Ready to connect★ 1k
- View details
git-and-commit-guidelinesSkillDev tools
Gives your agent rules for writing standardized commit messages and pull requests in the Trezor Suite repo.
Ready to connect★ 1k
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.
53,789 of the 54,221 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.