データサイエンティスト

ML Pipeline Skill

Builds reproducible, production-quality machine learning pipelines using scikit-learn's Pipeline and ColumnTransformer. Use this skill whenever the user needs to chain preprocessing and modeling steps together, prevent data leakage, handle mixed feature types (numeric + categorical), serialize a trained model, build a reusable ML workflow, or automate feature engineering within a pipeline. Trigger when the user says "build a pipeline", "automate my preprocessing", "I keep getting data leakage", "make this reproducible", "deploy this model", "save my model", "chain these steps together", or when you notice the user is fitting transformers on the full dataset before splitting (a leakage red flag). Also use when the user wants to do GridSearchCV across both preprocessing and model parameters simultaneously.

Phife7260

教育

Forecasting Tools

Metaculus forecasting-tools library for building prediction bots, fetching questions, and conducting research. Use when working with Metaculus questions, creating forecast bots, using SmartSearcher, or backtesting predictions.

rapturt90

営業・マーケティング

Multimodal Image Studio

当主要输入或输出涉及图像生成、图像编辑或图像风格迁移时，随时使用此技能。这意味着适用于用户希望执行以下操作的任何任务：根据简短文本描述或粗略草图生成新图像；对现有照片进行语义编辑（例如，添加对象、更改表情、修改背景）；或将参考图像的艺术风格迁移到内容图像上。特别是当用户提及绘制、生成、重塑或编辑视觉内容时触发——即使是随口提及（比如“把这家伙变黄”或“把我的照片变成3D盲盒”）。交付成果必须是图像文件。当主要交付成果是纯文本分析、文档创建或纯数据处理时，请勿触发此技能。

WuChaseSea0

Hotspots

用于拉取并展示最新热点内容（接口为 https://hotspot.api4claw.com/hotspots/latest?timestamp=$TIME_STEMP，返回 JSON 数组）。按模型预估点击率生成 Top 条目（不使用原始热度排序）。默认会提示用户创建定时任务（不自动执行），在用户确认后提供 cron 注册命令。适用于用户查询最新热点、服务状态、按 source_name 分组标题、Top 内容，或由已存在的定时消息触发执行。触发词："今日热点"、"今日头条"、"热点新闻"、"热点"

xltang0

データエンジニアリング

Consolidation Run

Execute the CAD consolidation pipeline (full mode), verify record counts against expected thresholds, and report quality metrics. Use when running a new consolidation after monthly data arrives.

racmac570

ラボツール

Wandb Tracker

Integrate Weights & Biases tracking for training observability, artifact lineage, and experiment comparison.

Rachasumanth0

パッケージ・配布

Mineagent Python

Python package layout for MineAgent: Gymnasium MinecraftEnv, AgentV1 (perception, affector, PPO, ICM, ROI), engine loop, and YAML config. Use when editing mineagent/, RL training logic, observation tensors, or run configuration.

thopkins320

データ分析

Using ba

Guide for using the ba (Bayesian Association) library to analyze categorical data. Use this skill whenever someone wants to explore associations in a dataset, compute contingency table metrics, run Bayesian inference on binary or categorical data, perform QCA (Qualitative Comparative Analysis), mine association rules, choose priors for small-sample analysis, or interpret Bayes factors and credible intervals. Also triggers on questions about odds ratios, relative risk, phi coefficient, lift, support, confidence, truth tables, Boolean minimization, or prior sensitivity analysis.

thorwhalen0

Previous 1…812 813 814…818 Next

Web・デジタルインターフェースデザイナー

コンピュータユーザーサポートスペシャリスト

Web・デジタルインターフェースデザイナー

コンピュータユーザーサポートスペシャリスト

Guidance

Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework

フレームワーク内部構造

Llama Cpp

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.

OBLITERATUS Skill

Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, LEACE, SAE decomposition, etc.) to excise guardrails while preserving reasoning. 9 CLI methods, 28 analysis modules, 116 model presets across 5 compute tiers, tournament evaluation, and telemetry-driven recommendations. Use when a user wants to uncensor, abliterate, or remove refusal from an LLM.

ゲーム開発

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.

Clip

OpenAI's model connecting vision and language. Enables zero-shot image classification, image-text matching, and cross-modal retrieval. Trained on 400M image-text pairs. Use for image search, content moderation, or vision-language tasks without fine-tuning. Best for general-purpose image understanding.

LLM・AI

Whisper

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.

Grpo Rl Training

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

Unsloth

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

Pytorch Fsdp

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

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.

Peft Fine Tuning

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Segment Anything Model

Foundation model for image segmentation with zero-shot transfer. Use when you need to segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks in an image.

CLI ツール

Jupyter Live Kernel

Use a live Jupyter kernel for stateful, iterative Python execution via hamelnb. Load this skill when the task involves exploration, iteration, or inspecting intermediate results — data science, ML experimentation, API exploration, or building up complex code step-by-step. Uses terminal to run CLI commands against a live Jupyter kernel. No new tools required.

LLM・AI

Dspy

Build complex AI systems with declarative programming, optimize prompts automatically, create modular RAG systems and agents with DSPy - Stanford NLP's framework for systematic LM programming

Traceable Experiment Loop

Use when running end-to-end experiments in scratchpad with strict traceability across scoping, planning, iterative build, migration, review, closeout, and final documentation review.

thomas-deboer0

データエンジニアリング

Data Loading Traceability

Use when implementing or updating data loading so every dataset source, transform, split, and filtering decision is documented and reproducible.

thomas-deboer0

金融・投資

scaleinvariance Package Reference

Use when working with scaleinvariance package for multifractal field simulation and analysis - Hurst exponent estimation, fractional Brownian motion (fBm), fractionally integrated flux (multiplicative cascades) (FIF), intermittency analysis, and Generalized Scale Invariance (GSI). Also use for scaling analysis of time series, turbulence data, atmospheric fields, or any task involving self-similar, cascade, or multifractal processes or fields.

thomasdewitt0

Ai Scientist

Experimental AI workflows, profiling constraints, computational acceleration strategies, and formula discovery algorithms. USE FOR: parameter searches, reinforcement learning for formula discovery, performance optimization, benchmarking, C extension writing, high-precision computation, symbolic computation with SymPy/Mathematica/RISC tools, formula harvesting from literature.

RamanujanMachine0

バイオインフォマティクス

Deseq2 Analysis

Pseudobulk DESeq2 differential expression for scRNA-seq. Use when running DE between conditions, comparing wound timepoints, or performing pseudobulk analysis.

rahulsarate0