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
54,220 results · page 453 of 1,808
- View details
comfyui-frontend-extensionsSkillDev tools
Lets your agent write ComfyUI frontend extension code like custom nodes, sidebar tabs, and widgets.
Ready to connect★ 748
- View details
comfyui-launch-flagsSkillDocs & knowledge
Helps your agent choose the right ComfyUI startup flags to fix out-of-memory errors and speed up image generation.
Ready to connect★ 748
- View details
comfyui-node-registrySkillDev tools
Lets your agent author and publish ComfyUI custom nodes to the Comfy Registry.
Ready to connect★ 748
- View details
debug-renderSkillMedia
Lets your agent debug a finished image or video render that looks wrong by inspecting each pipeline step.
Ready to connect★ 748
- View details
ernie-imageSkillMedia
Lets your agent generate images from text prompts, including posters, signage, and multi-panel layouts.
Ready to connect★ 748
- View details
flux-txt2imgSkillDev tools
Lets your agent build text-to-image workflows using Flux models and ComfyUI node patterns.
Ready to connect★ 748
- View details
ideogram-ultraSkillMedia
Lets your agent generate images from text or reference images using Ideogram Ultra workflows with strong text rendering.
Ready to connect★ 748
- View details
installer-packsSkillAI & models
Lets your agent install model packs and build new ones from installers or workflow JSON.
Ready to connect★ 748
- View details
krea2-identity-editSkillDev tools
Lets your agent swap clothing on a person in an image while keeping their face, pose and background.
Ready to connect★ 748
- View details
krea2-txt2imgSkillMedia
Lets your agent generate images from text prompts using fast 8-step Krea 2 Turbo workflows.
Ready to connect★ 748
- View details
local-llm-freeSkillAI & models
Lets your agent get help running AI models locally for free and offline with Ollama, no API key needed.
Ready to connect★ 748
- View details
lora-managerSkillAI & models
Helps your agent add LoRA Manager nodes in ComfyUI when standard node-adding fails.
Ready to connect★ 748
- View details
ltx-directorSkillMedia
Lets your agent edit LTX Director video timelines by adding, moving, and retiming clips and prompts.
Ready to connect★ 748
- View details
minimax-h3-videoSkillMedia
Lets your agent build local AI video generation workflows that create clips from text or images using ComfyUI nodes.
Ready to connect★ 748
- View details
model-compatibilitySkillMedia
Lets your agent look up which image and video models work with which loaders, samplers, resolutions, and adapters.
Ready to connect★ 748
- View details
model-registrySkillMedia
Gives your agent curated download links and folders for AI image and video models used with ComfyUI.
Ready to connect★ 748
- View details
panel-node-pack-syncSkillAI & models
Keeps the ComfyUI sidebar panel node-pack aligned with the orchestrator after comfyui-mcp updates.
Ready to connect★ 748
- View details
panel-operationsSkillProductivity
Gives your agent step-by-step procedures for building and editing ComfyUI workflows in the sidebar panel.
Ready to connect★ 748
- View details
qwen-image-editSkillMedia
Lets your agent set up image editing workflows with Qwen, including model loading, prompts, and test grids.
Ready to connect★ 748
- View details
qwen-txt2imgSkillMedia
Lets your agent generate images from text descriptions using Qwen image models.
Ready to connect★ 748
- View details
report-bugSkillFiles & storage
Lets your agent explain that comfyui-mcp is archived and direct users to ComfyUI's official tooling instead of filing bug reports.
Ready to connect★ 748
- View details
rgthreeSkillDev tools
Lets your agent configure rgthree ComfyUI workflow nodes like group toggles, LoRA loaders, seeds, and switches.
Ready to connect★ 748
- View details
train-character-loraSkillDatabases & data
Lets your agent train a custom image model of a person or character from your photos on your local GPU.
Ready to connect★ 748
- View details
triton-sageattentionSkillMedia
Speeds up ComfyUI video generation by installing Triton and SageAttention acceleration into the right Python setup on Windows or Linux.
Ready to connect★ 748
- View details
troubleshootingSkillMedia
Lets your agent diagnose and fix common ComfyUI errors like out-of-memory, missing nodes, and black images.
Ready to connect★ 748
- View details
video-upscaleSkillMedia
Lets your agent upscale and restore low-resolution videos using AI models in ComfyUI.
Ready to connect★ 748
- View details
wan-flf-videoSkillMedia
Lets your agent generate videos from a first and last frame using WAN 2.2 workflows in ComfyUI.
Ready to connect★ 748
- View details
wan-multitalkSkillMedia
Lets your agent create lip-synced talking-avatar videos from a portrait image and an audio clip.
Ready to connect★ 748
- View details
wan-scail-replacementSkillMedia
Lets your agent replace characters in videos using SCAIL-2 on WAN 2.1 with reference-image framing and masking.
Ready to connect★ 748
- View details
wan-t2v-videoSkillMedia
Lets your agent generate videos from text descriptions using WAN 2.2 workflows with two-pass sampling.
Ready to connect★ 748
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.
53,788 of the 54,220 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.