MLOps
模型部署、评估与运维
pyvene-interventions
Orchestra-Research
Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior.
evaluating-llms-harness
Orchestra-Research
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
unsloth
Orchestra-Research
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
gguf-quantization
Orchestra-Research
GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.
transformer-lens-interpretability
Orchestra-Research
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.
grpo-rl-training
Orchestra-Research
Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training
sentencepiece
Orchestra-Research
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.
rwkv-architecture
Orchestra-Research
RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.
distributed-llm-pretraining-torchtitan
Orchestra-Research
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.
lstm-price-prediction-guide
nirholas
Guide to LSTM-based cryptocurrency price prediction. Covers data preprocessing, feature engineering, LSTM architecture, training, evaluation, and deployment. Includes TensorFlow/Keras implementation with technical indicators and sentiment features.
giza-zkml-visualization
nirholas
Giza ZKML Agent Visualization — guide covering key concepts, implementation patterns, and best practices.
skill-model-updater
bitwize-music-studio
Updates model references across all skill files when new Claude models are released. Use when Anthropic releases new Claude models to keep skills current.
mix-engineer
bitwize-music-studio
Polishes raw Suno audio by processing per-stem WAVs (vocals, drums, bass, other) with targeted cleanup, EQ, and compression, then remixing into a polished stereo WAV ready for mastering. Use after audio import and before mastering.
about
bitwize-music-studio
Provides information about the bitwize-music plugin and its creator. Use when the user asks about the plugin, its purpose, or capabilities.
nnsight-remote-interpretability
Orchestra-Research
Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture.
django-perf-review
getsentry
Django performance code review. Use when asked to "review Django performance", "find N+1 queries", "optimize Django", "check queryset performance", "database performance", "Django ORM issues", or audit Django code for performance problems.
firebase-ai-logic
firebase
Official skill for integrating Firebase AI Logic (Gemini API) into web applications. Covers setup, multimodal inference, structured output, and security.
nano-banana
Roasbeef
Generate and edit images using Gemini's image generation API (Imagen 3). This skill should be used when creating images, illustrations, diagrams, editing existing images, or iteratively refining visual content through multi-turn conversations.
Video Editing
6missedcalls
Output: always tell the user the output file path and duration
ols-regression
brycewang-stanford
Econometrics skill for OLS regression and linear models. Activates when the user asks about: "run OLS", "linear regression", "ordinary least squares", "interpret regression results", "heteroskedasticity", "multicollinearity", "regression assumptions", "robust standard errors", "GLS", "WLS", "fit a regression model", "check regression diagnostics", "OLS假设", "最小二乘法", "线性回归", "回归系数", "残差检验", "异方差", "多重共线性", "普通最小二乘", "稳健标准误", "回归诊断"
novel-summarizer
GongLingRui
阅读理解小说文本，总结成流畅的故事大纲。适用于小说初筛选、生成500-800字故事大纲
novel-evaluator
GongLingRui
严格细致判断与评分故事文本，从市场潜力、创新属性、内容亮点维度分析质量。适用于小说初筛选、多维度评估打分
script-evaluator
GongLingRui
从思想性、艺术性、观赏性三维度评估影视剧本并打分。适用于剧本开发质量评估、修改方向确定、项目立项前审查
dbs-diagnosis
dontbesilent2025
dontbesilent 商业模式诊断。两种模式：问诊（消解你的问题）和体检（拆解你的商业模式）。 触发方式：/dbs-diagnosis、/问诊、「帮我看看商业模式」「诊断一下我的业务」「我有个商业问题」 Business model diagnosis using dontbesilent's ontological framework. Two modes: consultation (dissolve your question) and checkup (analyze your business model). Trigger: /dbs-diagnosis, "diagnose my business model", "I have a business question"