Find the right skill for the job.

skill-creator

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

algorithmic-art

Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.

1code

Imported from https://www.skillhub.club/skills/21st-dev-1code.

model-pruning

Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50% sparsity with minimal accuracy loss, or enabling faster inference on hardware accelerators. Covers unstructured pruning, structured pruning, N:M sparsity, magnitude pruning, and one-shot methods.

Emerging TechniquesModel PruningWandaSparseGPT

miles-rl-training

Provides guidance for enterprise-grade RL training using miles, a production-ready fork of slime. Use when training large MoE models with FP8/INT4, needing train-inference alignment, or requiring speculative RL for maximum throughput.

Reinforcement LearningMoEFP8INT4

grpo-rl-training

Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training

Post-TrainingReinforcement LearningGRPOTRL

sentencepiece

Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or reproducible tokenization.

TokenizationSentencePieceLanguage-IndependentBPE

deepspeed

Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention

DeepSpeedDistributed TrainingZeROPipeline Parallelism

audiocraft-audio-generation

PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen). Use when you need to generate music from text descriptions, create sound effects, or perform melody-conditioned music generation.

MultimodalAudio GenerationText-to-MusicText-to-Audio

fine-tuning-with-trl

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.

Post-TrainingTRLReinforcement LearningFine-Tuning

distributed-llm-pretraining-torchtitan

Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.

Model ArchitectureDistributed TrainingTorchTitanFSDP2

long-context

Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques. Use when processing long documents (32k-128k+ tokens), extending pre-trained models beyond original context limits, or implementing efficient positional encodings. Covers rotary embeddings, attention biases, interpolation methods, and extrapolation strategies for LLMs.

Emerging TechniquesLong ContextRoPEYaRN

openrlhf-training

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.

Post-TrainingOpenRLHFRLHFPPO

devportfolio

Imported from https://www.skillhub.club/skills/ryanfitzgerald-devportfolio.

code-review-small

Quick code review for small PRs (<100 lines). Focuses on basic code quality, naming conventions, and obvious bugs.

control-flow

Human-readable control flow patterns for refactoring complex conditionals. Use when refactoring nested conditionals, improving code readability, or restructuring decision logic.

algorithmic-art

kubefwd

Imported from https://www.skillhub.club/skills/txn2-kubefwd.

ccg-workflow

Imported from https://www.skillhub.club/skills/fengshao1227-ccg-workflow.

create-changelog-announcement

Use this skill to create and publish changelog announcements for new features, improvements, or bug fixes. This skill handles the complete workflow - creating detailed changelog documentation pages, adding sidebar announcement cards, and ensuring everything follows project standards. Use when the user mentions adding changelog entries, documenting new features, creating release notes, or announcing product updates.