Data Science & ML Claudeスキル

NebulaOS LLM Gateway

Manages NebulaOS LLM infrastructure including provider routes, API keys, model catalogs, and cost monitoring.

Advanced Evaluation (LLM-as-Judge)

Implements high-rigor evaluation frameworks with bias mitigation and pairwise comparisons for mission-critical AI outputs.

Cost-Aware LLM Pipeline

Optimizes LLM API expenditures through intelligent model routing, budget guardrails, and efficient prompt caching patterns.

Marimo Notebook Essentials

Develops reactive Python notebooks and interactive dashboards using the marimo framework.

Economic Calendar Fetcher

Fetches and analyzes global economic events and market-moving indicators using the Financial Modeling Prep API.

Bakken Run Coach

Analyzes fitness training data to provide professional running coaching feedback based on the Bakken Norwegian Model.

Advanced Excel & Spreadsheet Automation

Manages complex spreadsheet creation, data analysis, and formula-driven financial modeling with high precision.

Advanced Excel & Financial Modeling

Automates the creation, editing, and analysis of professional Excel workbooks with dynamic formulas and industry-standard financial modeling.

Scverse Single-Cell Analysis

Streamlines single-cell RNA-seq analysis using standardized AnnData and Scanpy workflows.

Plotly Compact

Generates high-performance interactive visualizations using Plotly Express and Graph Objects patterns.

Context Engineering

Optimizes LLM context windows to maximize reasoning quality while minimizing token costs and latency.

Modern R Development

Implements modern R programming patterns focusing on tidyverse efficiency, metaprogramming, and performance-optimized package development.

scverse & Scanpy Essentials

Streamlines single-cell transcriptomics workflows using standardized scverse, AnnData, and Scanpy implementation patterns.

Snakemake Compact Workflows

Streamlines Snakemake pipeline development using modular patterns and compact rule definitions for reproducible data science.

Phosphoproteomics PTM Analysis

Analyzes phosphoproteomics data using R to identify differential post-translational modification abundance, stoichiometry, and kinase activity.

Sanctuary DNA Agent DSL

Constructs sophisticated Gnostic agent workflows using a structured DSL built on LangGraph's execution substrate.

AI Engineering & MLOps Guide

Provides comprehensive guidance and best practices for designing, building, and scaling production-grade AI systems and machine learning pipelines.

ClinicalTrials.gov Search

Queries the ClinicalTrials.gov API v2 to retrieve comprehensive data on clinical studies worldwide for research, patient matching, and data analysis.

NeuroScript Standard Library

Navigates and documents the NeuroScript standard library for composing complex neural architectures.

Content-Hash Cache Pattern

Implements an efficient, path-independent caching system for expensive file processing tasks using SHA-256 content hashes.

STRING Protein Interaction Analysis

Queries the STRING database to analyze protein-protein interaction networks and perform functional enrichment for systems biology research.

ClinicalTrials.gov Search & Analytics

Queries the ClinicalTrials.gov API v2 to retrieve, filter, and export detailed clinical study data for research and patient matching.

R Economics Code Review

Performs automated code reviews on R scripts and Quarto documents to ensure reproducibility and statistical correctness in economics research.

ZINC Database Search

Accesses and searches the ZINC22 database of 230M+ purchasable compounds for virtual screening and lead discovery.

STRING Database Analysis

Analyzes protein-protein interaction networks and performs functional enrichment using the STRING database REST API.

Synthea FHIR Test Data Generator

Generates high-fidelity synthetic FHIR patient data from PheKB clinical phenotype definitions using Synthea and UMLS.

AG2 Agent Scaffolder

Generates structured AG2 ConversableAgents with integrated tool functions and optimized LLM configurations.

MbTorch I/O & Model Formats

Manages high-performance model I/O for the MbTorch framework, supporting ONNX, safetensors, and native binary formats with strict validation.

AG2 Group Chat AutoPattern

Implements autonomous multi-agent group chats using LLM-driven agent selection for dynamic and unpredictable conversation flows.

Haipipe Project Manager

Standardizes and automates project scaffolding, organization, and documentation for machine learning pipelines within the haipipe workspace.