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 114 of 1,808
- View details
modal-serverless-gpuSkillCloud & infra
Lets your agent run ML jobs on cloud GPUs, deploying models as APIs and scaling automatically.
Ready to connect★ 13k
- View details
model-mergingSkillAI & models
Lets your agent combine multiple fine-tuned AI models into one, blending skills without retraining.
Ready to connect★ 13k
- View details
model-pruningSkillAI & models
Guides your agent through pruning large language models to shrink their size and speed up inference without retraining.
Ready to connect★ 13k
- View details
moe-trainingSkillAI & models
Guides your agent through building and training Mixture of Experts AI models with DeepSpeed or HuggingFace.
Ready to connect★ 13k
- View details
nanogptSkillAI & models
Lets your agent write, train, and sample from a minimal GPT model in Python to learn how transformers work.
Ready to connect★ 13k
- View details
nemo-curatorSkillDatabases & data
Lets your agent clean and prepare large training datasets by filtering, deduplicating, and redacting text, image, video, and audio data.
Ready to connect★ 13k
- View details
nemo-guardrailsSkillAI & models
Lets your agent check and filter LLM inputs and outputs for jailbreaks, hallucinations, PII, and toxicity.
Ready to connect★ 13k
- View details
nnsight-remote-interpretabilitySkillAI & models
Lets your agent inspect and experiment with the internal activations of neural networks, running large models remotely.
Ready to connect★ 13k
- View details
openrlhf-trainingSkillAI & models
Lets your agent train large language models with reinforcement learning methods like PPO and DPO using distributed GPUs.
Ready to connect★ 13k
- View details
optimizing-attention-flashSkillDocs & knowledge
Guides your agent to speed up transformer attention in PyTorch code using Flash Attention for faster, lower-memory training and inference.
Ready to connect★ 13k
- View details
outlinesSkillAI & models
Makes your agent's model outputs always follow a valid JSON, XML, or code structure.
Ready to connect★ 13k
- View details
peft-fine-tuningSkillDocs & knowledge
Guides your agent to fine-tune large language models with LoRA and QLoRA using little GPU memory.
Ready to connect★ 13k
- View details
phoenix-observabilitySkillDatabases & data
Lets your agent trace and debug LLM app runs, run evaluations on datasets, and monitor AI systems in real time.
Ready to connect★ 13k
- View details
pineconeSkillSearch
Lets your agent store and search text by meaning for fast semantic search and RAG at scale.
Ready to connect★ 13k
- View details
presenting-conference-talksSkillDocs & knowledge
Turns a research paper into conference presentation slides with speaker notes and a talk script.
Ready to connect★ 13k
- View details
prompt-guardSkillCloud & infra
Checks incoming prompts and third-party text for injection attacks or jailbreaks before your agent acts on them.
Ready to connect★ 13k
- View details
pytorch-fsdp2SkillDocs & knowledge
Helps your coding agent add PyTorch FSDP2 sharding to training scripts so large models can train across multiple GPUs.
Ready to connect★ 13k
- View details
pyvene-interventionsSkillMonitoring & ops
Lets your agent run causal interventions like activation patching and causal tracing on PyTorch models using pyvene.
Ready to connect★ 13k
- View details
qdrant-vector-searchSkillSearch
Lets your agent search stored documents by meaning to find relevant information for answering questions.
Ready to connect★ 13k
- View details
quantizing-models-bitsandbytesSkillDocs & knowledge
Helps your agent load and fine-tune large language models in 8-bit or 4-bit so they fit in limited GPU memory.
Ready to connect★ 13k
- View details
ray-dataSkillDatabases & data
Lets your agent process large datasets for machine learning using distributed pipelines with Parquet, CSV, JSON, and images.
Ready to connect★ 13k
- View details
ray-trainSkillAI & models
Lets your agent run machine learning training across many machines, scaling PyTorch or HuggingFace models with automatic tuning.
Ready to connect★ 13k
- View details
rwkv-architectureSkillProductivity
Guides your agent in running, fine-tuning, and evaluating RWKV models, an efficient RNN-Transformer hybrid with linear-time inference.
Ready to connect★ 13k
- View details
segment-anything-modelSkillMedia
Lets your agent segment objects in images using points or boxes, or automatically generate masks for every object.
Ready to connect★ 13k
- View details
sentence-transformersSkillSearch
Lets your agent turn text or images into embeddings for semantic search, clustering, and similarity tasks.
Ready to connect★ 13k
- View details
sentencepieceSkillDocs & knowledge
Lets your agent split text into tokens for NLP tasks using fast language-independent BPE or Unigram tokenization.
Ready to connect★ 13k
- View details
serving-llms-vllmSkillDocs & knowledge
Guides your agent through deploying and running LLM models efficiently with vLLM, including serving APIs and quantization.
Ready to connect★ 13k
- View details
sglangSkillAI & models
Lets your agent run its own local LLM server that generates JSON or regex-structured output quickly with prefix caching.
Ready to connect★ 13k
- View details
simpo-trainingSkillAI & models
Lets your agent fine-tune a language model to match preferences using a simpler, faster method than DPO.
Ready to connect★ 13k
- View details
skypilot-multi-cloud-orchestrationSkillDev tools
Lets your agent launch and manage machine learning training jobs across multiple cloud providers while picking the cheapest GPUs.
Ready to connect★ 13k
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.