Skill UI

A comprehensive guide for defining, tracking, and interpreting product metrics across all stages—from discovery to maturity. This skill covers selecting metric frameworks (AARRR, HEART), defining stage-appropriate KPIs, designing multi-layered dashboards, and executing deep analyses like cohort and retention modeling. It helps teams transform raw data into actionable product strategies, moving beyond vanity metrics.

This skill systematically verifies the novelty of a proposed research method or idea against vast global literature. It identifies core technical claims, performs multi-source searches across arXiv, Google Scholar, and major conferences (ICLR, NeurIPS, ICML), and utilizes advanced LLMs for cross-model verification. The output is a structured report, providing a novelty score, identifying the closest prior work, and offering a strategic recommendation to guide the research direction.

bdistill transforms AI interactions into a structured, queryable knowledge base. It extracts, structures, and quality-scores domain knowledge from both closed-source AI models (in-session) and local open-source models (via Ollama). Users can build comprehensive reference datasets, Q&A pairs, or generate structured training data for traditional machine learning, ensuring validated and exportable domain expertise.

Collection of battle-tested PyTorch training recipes covering architecture selection, optimizers, LR scheduling, precision tips, and debugging to train or fine-tune LLMs, vision, diffusion, and biomedical models with high GPU throughput.

This skill provides comprehensive command-line interface (CLI) functionality for managing the entire lifecycle of experiments within the Arize platform. Users can list, retrieve metadata, create new experiments from various data files (CSV, JSON, Parquet), and export detailed run results for deep analysis. It supports linking experiments to specific dataset versions, making it essential for structured ML model evaluation and performance comparison workflows.

This skill provides comprehensive command-line capabilities for interacting with Arize datasets. Use it to perform dataset CRUD operations (Create, Read, Update), list available datasets, retrieve detailed metadata, and export large volumes of example data. It is essential for managing the lifecycle of versioned datasets used in model evaluation and machine learning experimentation. Supports file formats like CSV, JSON, and Parquet.

Structured pipeline for credit risk pre-loan modeling that loads and formats raw data, filters abnormal periods, trims high-missing/low-IV/high-PSI/high-correlation features, denoises via Null Importance, and exports an Excel cleaning report for review.

Surfaces Monte Carlo data observability insights—health, lineage, alerts, and monitors-as-code—before developers edit dbt models or SQL pipelines, ensuring risk-aware code changes.

This skill automatically generates an interactive SQL Notebook designed to validate changes in dbt models or snapshots. It processes changes identified from a GitHub PR or local repository, generating comparison queries that compare 'before' (production) and 'after' (development) data states. It is essential for embedding data observability and quality checks into CI/CD pipelines.

Guides a Power BI DAX expert through performance, readability, best-practice, and maintainability reviews; recommends variable reuse, context tuning, and error handling; then outputs an optimized formula plus justification for analytics models.

A comprehensive expert prompt for evaluating Power BI data models. It guides the assessment of model architecture (Star Schema, relationships), storage optimization (Import, DirectQuery, Dual), data quality, performance scalability, and governance best practices. Use this when reviewing an existing BI model to ensure it is robust, maintainable, and highly performant.

Provides systematic Power BI performance troubleshooting guidance covering model optimization, DAX design, report layout, and capacity diagnostics, plus tool-driven measurement tips for faster page loads, interactions, and query execution.