Data Science & ML Claudeスキル

Sibyl LaTeX Writer

Converts research papers into NeurIPS LaTeX format and automates the document compilation process.

Sibyl Supervisor Decision

Analyzes AI experiment results and expert debates to determine whether a project should pivot or proceed.

Sibyl Contrarian

Challenges mainstream assumptions and identifies project blind spots by uncovering counter-evidence through deep academic and web research.

Sibyl Final Critic

Provides rigorous, conference-grade peer reviews for machine learning research papers targeting NeurIPS and ICML standards.

Sibyl Comparative Analyst

Conducts deep comparative research against State-of-the-Art (SOTA) works to define and position project contributions.

Sibyl Pragmatist Research Agent

Generates engineering-focused research proposals by validating academic literature against practical implementation constraints.

Sibyl Experiment Supervisor

Orchestrates machine learning experiments by monitoring GPU availability and dynamically dispatching tasks to remote infrastructure.

Sibyl Literature Researcher

Conducts systematic literature reviews by synthesizing data from arXiv, Google Scholar, bioRxiv, and general web searches.

Sibyl Server Experimenter

Orchestrates and executes machine learning experiments on remote servers using local Claude or Codex agents.

Sibyl Empiricist

Facilitates rigorous experimental design and ensures data reproducibility by integrating academic research tools into AI workflows.

Weaviate Vector Database

Streamlines AI-native application development by providing expert patterns for Weaviate vector search, RAG pipelines, and multi-tenancy.

Perplexity Sonar API Integration

Integrates real-time web search and grounded AI responses into applications using the Perplexity Sonar API.

OpenAI SDK Integration

Simplifies the development of AI-powered applications using the OpenAI SDK and the GPT-5 model family.

LangGraph Agent Orchestration

Builds stateful, multi-agent applications and cyclic AI workflows using LangGraph for complex orchestration and persistent state management.

Opus 4.5 Model Migrator

Migrates codebases and prompts from earlier Claude models to Opus 4.5 by automating model string updates and behavioral adjustments.

Instinct Extractor

Analyzes raw interaction logs to extract recurring behavioral patterns and refine AI decision-making.

Advanced Prompt Architect

Generates and optimizes high-performance prompts for Claude, GPT, and other LLMs using research-backed engineering techniques.

Paper Council

Transforms research papers and code repositories into high-quality, reusable AI skills through a multi-agent peer-review process.

Academic Research Paper Finder

Finds and prioritizes academic research papers to streamline literature reviews and citation management directly within your workspace.

AI Multimodal Processing

Enables Claude to analyze, transcribe, and generate multimedia content including audio, images, videos, and documents through the Gemini API.

Quantitative Finance & Strategy Analyst

Builds sophisticated financial models and backtests algorithmic trading strategies with integrated risk management metrics.

PyHealth AI Toolkit

Develops, tests, and deploys healthcare AI models using clinical data and specialized medical coding systems.

Economics Data Analysis

Automates end-to-end empirical data analysis workflows using Stata and R for publication-ready research output.

RAP Compliance Auditor for R

Audits R projects against UK Government Reproducible Analytical Pipelines (RAP) maturity standards to ensure code quality and transparency.

DiffDock Molecular Docking

Performs diffusion-based molecular docking to predict high-accuracy 3D binding poses between proteins and ligands for drug discovery.

Bayesian Modeling in R

Builds and analyzes robust Bayesian statistical models using Stan-based R packages like brms and rstanarm.

Economic Data Statistics Methods

Calculates advanced economic statistics and estimates using specialized R methodologies for weighted and bunched data.

Gemini Veo Video Generation

Provides comprehensive technical specifications and implementation patterns for Google's Veo 3.1 video generation model.

Robot Learning Run Evaluator

Analyzes, summarizes, and documents robot learning experiment runs across local and cluster environments.

Llama.cpp Runtime Controller

Manages local GGUF model inference and API serving through llama.cpp within worker terminals.