MLOps
模型部署、评估与运维
detailed-plot-analyzer
GongLingRui
基于大情节点，深入分析并生成详细情节点描述及情节发展说明。适用于细化故事大纲、指导剧本写作、分析情节内在逻辑
story-outliner
GongLingRui
阅读理解故事文本，总结人物、关系、情节，整理成流畅大纲。适用于快速了解故事核心、为剧本创作提供大纲基础
story-outline-evaluator
GongLingRui
专业评估故事大纲，从市场潜力、创新属性、内容亮点维度判断评分。适用于故事大纲质量评估、IP改编潜力判断、项目立项决策
story-summarizer
GongLingRui
基于故事文本提炼主要情节与要点，生成完整故事梗概。适用于快速了解故事内容、剧本改编、项目推介
drama-planner
GongLingRui
制定竖屏短剧专业策划方案，含情绪价值分析、黄金三秒钩子设计、三幕式结构规划等。适用于短剧项目前期策划、商业化方案设计、创作指导
plot-points-analyzer
GongLingRui
分析故事情节点，识别关键情节与转折点。适用于深度分析情节结构、评估情节发展有效性
dowhy
brycewang-stanford
Causal inference framework for answering "does X cause Y?" beyond correlation. DoWhy (Microsoft Research) provides the identify-estimate-refute loop: define a causal graph (DAG), identify the causal effect using backdoor/frontdoor/instrumental variable criteria, estimate treatment effects with multiple estimators, and validate results with automated refutation tests. Use when: distinguishing causation from correlation, estimating treatment effects (ATE, ATT, CATE), designing and analyzing A/B tests with confounders, using instrumental variables, performing counterfactual reasoning ("what would have happened if..."), validating causal claims with sensitivity analysis, working with observational data where randomization is impossible, or any analysis where the question is "what is the CAUSAL effect of X on Y" rather than just "how do X and Y relate?"
PyFixest LLM Skill Reference
brycewang-stanford
```
causal-inference-root-cause
lyndonkl
Use when investigating why something happened and need to distinguish correlation from causation, identify root causes vs symptoms, test competing hypotheses, control for confounding variables, or design experiments to validate causal claims. Invoke when debugging systems, analyzing failures, researching health outcomes, evaluating policy impacts, or when user mentions root cause, causal chain, confounding, spurious correlation, or asks "why did this really happen?"
ai-engineer
rmyndharis
Build production-ready LLM applications, advanced RAG systems, and
ai-engineer
sickn33
Build production-ready LLM applications, advanced RAG systems, and
LLM Council
dair-ai
Orchestrate multiple open-weight LLMs via Fireworks AI to deliberate on queries. Models respond individually, rank each other's responses, then a Chairman synthesizes the final answer. Use this skill when the user wants multiple AI perspectives, consensus-building, or the "LLM Council" approach inspired by Karpathy. Powered by fast, affordable open-weight models on Fireworks.
ray-data
Orchestra-Research
Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images. Integrates with Ray Train, PyTorch, TensorFlow. Scales from single machine to 100s of nodes. Use for batch inference, data preprocessing, multi-modal data loading, or distributed ETL pipelines.
awq-quantization
Orchestra-Research
Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.
pytorch-lightning
Orchestra-Research
High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.
training-llms-megatron
Orchestra-Research
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.
fine-tuning-with-trl
Orchestra-Research
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.
ML Training Recipes
Orchestra-Research
Comprehensive open-source library of AI research and engineering skills for any AI model. Package the skills and your claude code/codex/gemini agent will be an AI research agent with full horsepower. Maintained by Orchestra Research.
sparse-autoencoder-training
Orchestra-Research
Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.
lambda-labs-gpu-cloud
Orchestra-Research
Reserved and on-demand GPU cloud instances for ML training and inference. Use when you need dedicated GPU instances with simple SSH access, persistent filesystems, or high-performance multi-node clusters for large-scale training.
huggingface-tokenizers
Orchestra-Research
Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.
llama-cpp
Orchestra-Research
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.
nanogpt
Orchestra-Research
Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU).
evaluating-code-models
Orchestra-Research
Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality. Industry standard from BigCode Project used by HuggingFace leaderboards.