Find the best CLAUDE.md, AGENTS.md and Claude skills

Clay-HHKSkill

sglang

Clay-HHK/claude-skills/sglang/SKILL.md

Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with tool calls, or when you need 5× faster inference than vLLM with prefix sharing. Powers 300,000+ GPUs at xAI, AMD, NVIDIA, and LinkedIn.

Clay-HHKSkill

simpo-training

Clay-HHK/claude-skills/simpo/SKILL.md

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.

Clay-HHKSkill

skill-creator

Clay-HHK/claude-skills/skill-creator/SKILL.md

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.

Often used withtaskflow-inbox-triage·mermaid-diagram-generator·ecosystem-reference

Clay-HHKSkill

Skill Development

Clay-HHK/claude-skills/skill-development/SKILL.md

This skill should be used when the user wants to "create a skill", "add a skill to plugin", "write a new skill", "improve skill description", "organize skill content", or needs guidance on skill structure, progressive disclosure, or skill development best practices for Claude Code plugins.

Often used withCommand Development·Plugin Settings·Hook Development

Clay-HHKSkill

skill-improver

Clay-HHK/claude-skills/skill-improver/SKILL.md

This skill should be used when the user asks to "apply skill improvements", "update skill from plan", "execute improvement plan", "fix skill issues", "implement skill recommendations", or mentions applying improvements from quality review reports. Reads improvement-plan-{name}.md files generated by skill-quality-reviewer and intelligently merges and executes the suggested changes to improve Claude Skills quality.

Clay-HHKSkill

skill-quality-reviewer

Clay-HHK/claude-skills/skill-quality-reviewer/SKILL.md

This skill should be used when the user asks to "analyze skill quality", "evaluate this skill", "review skill quality", "check my skill", or "generate quality report". Evaluates local skills across description quality, content organization, writing style, and structural integrity.

Clay-HHKReads credentialsSkill

skypilot-multi-cloud-orchestration

Clay-HHK/claude-skills/skypilot/SKILL.md

Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.

Clay-HHKSkill

slack-gif-creator

Clay-HHK/claude-skills/slack-gif-creator/SKILL.md

Knowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack.

Often used withacademy-guide·skill-share·raffle-winner-picker

Clay-HHKSkill

slime-rl-training

Clay-HHK/claude-skills/slime/SKILL.md

Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation workflows, or needing tight Megatron-LM integration for RL scaling.

Clay-HHKSkill

speculative-decoding

Clay-HHK/claude-skills/speculative-decoding/SKILL.md

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

Clay-HHKSkill

stable-diffusion-image-generation

Clay-HHK/claude-skills/stable-diffusion/SKILL.md

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.

Clay-HHKSkill

start-my-day

Clay-HHK/claude-skills/start-my-day/skill.md

论文阅读工作流启动 - 生成今日论文推荐笔记

Clay-HHKSkill

template-skill

Clay-HHK/claude-skills/template/SKILL.md

Replace with description of the skill and when Claude should use it.

Often used withskill-share·aesthetic·artifacts-builder

Clay-HHKSkill

tensorboard

Clay-HHK/claude-skills/tensorboard/SKILL.md

Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance with TensorBoard - Google's ML visualization toolkit

Clay-HHKSkill

tensorrt-llm

Clay-HHK/claude-skills/tensorrt-llm/SKILL.md

Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.

Clay-HHKSkill

theme-factory

Clay-HHK/claude-skills/theme-factory/SKILL.md

Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.

Often used withskill-share·context7-efficient·academy-guide

Clay-HHKSkill

torchforge-rl-training

Clay-HHK/claude-skills/torchforge/SKILL.md

Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.

Clay-HHKSkill

distributed-llm-pretraining-torchtitan

Clay-HHK/claude-skills/torchtitan/SKILL.md

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.

Clay-HHKSkill

transformer-lens-interpretability

Clay-HHK/claude-skills/transformer-lens/SKILL.md

Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.

Clay-HHKSkill

fine-tuning-with-trl

Clay-HHK/claude-skills/trl-fine-tuning/SKILL.md

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.