數據科學

Excel spreadsheet toolkit for creating, reading, and manipulating .xlsx files. Supports formulas, formatting, charts, and financial modeling with industry-standard conventions. Use for data analysis, financial models, reports, and spreadsheet automation.

Analyze database queries and recommend missing indexes with performance impact analysis

use when adding API to this project that exists in numpy, especially if the user mentions a stream in the numpy-api-parity plan.

Execute high-density architectural analysis on user ideas. Move from 'Vague' to 'Verified' using a 5-step logic chain: Calibration → Decomposition → Excavation → Re-Architecting → Inversion. This skill should be used when analyzing system architecture, validating technical ideas, or performing pre-mortems on solutions.

Standards for rigorous data analysis using OSEMN methodology. Focuses on statistical validation, model reliability, and AI-readability.

Analyzes business models including revenue models, unit economics, competitive moats, scalability, and value creation/capture mechanisms using frameworks like Business Model Canvas and strategic analysis. Use when the user requests business model analysis, unit economics review, moat assessment, or wants to understand how a company creates and captures value.

Expert business analysis for B2B SaaS platforms. Activated for data analysis, requirements gathering, process optimization, business metrics calculation, ROI analysis, and business case development.

Synchronized documentation updates across CLAUDE.md, README.md, .claude/CLAUDE.md, and deployment checklist to ensure consistency. Use when adding new architecture documents, analysis documents, or critical reference documents.

SPARC Pseudocode phase specialist for algorithm design, data structure selection, complexity analysis, and design pattern identification

Deploy and operate the vehicle insurance data analysis platform. Use when user asks about local development setup, production deployment, server configuration, build process, service management, or troubleshooting deployment issues. Focuses on the project's actual simple deployment model using start_server.sh, not complex enterprise setups.

Extended thinking (ultrathink) configuration for Claude API. Activate for complex reasoning tasks, deep analysis, multi-step problem solving, and tasks requiring careful deliberation. Enables Claude's internal reasoning with configurable thinking budgets.

Extract subtitles, frames, and metadata from YouTube videos. Use when user shares a YouTube URL and wants transcript, screenshots, or video analysis.

Comprehensive guide to phylogenetic tree building methods including distance-based (UPGMA, Neighbor-Joining), maximum likelihood (RAxML, IQ-TREE), and Bayesian inference (MrBayes). Covers multiple sequence alignment, distance matrices, bootstrap analysis, consensus methods, tree formats (Newick, Nexus), and tree comparison metrics. Includes implementation patterns for tree visualization and evaluation.

Fetch actual data from Danmarks Statistik API and store in DuckDB. Use when user wants to download and store specific DST table data for analysis.

Guidance for working with Awkward Array 2.0 jagged arrays and records in Python. Use when building or debugging `awkward` workflows, including record construction with `ak.zip`, adding fields with `ak.with_field`, filtering/aggregation, combinatorics (`ak.cartesian`/`ak.combinations`), `argmin`/`argmax` slicing, flattening, sorting, and NumPy interop or common Awkward pitfalls.

Master data analysis, visualization, and insights extraction.

Coordinates complete comparison of 2 products with parallel subagents and result aggregation. Use when user asks to 'compare X vs Y', 'compare two products', 'research products for comparison', mentions comparing items for purchase decision, or requests product analysis. Handles product validation, cache checking, parallel data collection (specs, reviews, pricing), and comprehensive report generation.

Test-first development strategy for PAI projects. USE WHEN user needs test strategy, coverage analysis, ATDD workflows, risk-based testing, or quality gates. Ensures tests are written before code, not after bugs appear.

Generate intelligent Python project documentation with UML class diagrams, architecture diagrams, and code analysis. Analyzes codebase structure to recommend and create appropriate visualizations for project overview.

Output format markers for the real-time stream formatter. Use when building prompts for streaming analysis to ensure proper progress display. Documents the patterns that StreamFormatter detects and displays.