菜单
First look at a dataset — profiles structure, data quality, and key distributions to build a mental model before any specific analysis. Use when the user wants to understand a new table or asks for an overview without a defined goal.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Sequence cleaning steps in a chain — trim/case/replace, null handling, deduplication, and type casting — in the right order. Use when raw ingested data needs cleaning before joins or aggregation.
Place Assert and Schema Validation nodes as gates that halt a chain when data is wrong, so bad data never reaches the output. Use when the pipeline must guarantee correctness before exporting or loading downstream.
Choose the right sink format, compression, and destination for a chain's output, and decide between exporting a file vs creating a table. Use at the end of a pipeline when deciding how to persist results.
Best practices for loading files into a chain — single file, folder globs, type detection, and union of many files. Use when the pipeline starts from CSV/Parquet/JSON/Excel files or a folder of files.
Decide between Merge (stack rows / UNION) and Join (match on a key) when a chain has multiple inputs, and set keys and join type correctly. Use when a pipeline combines two or more upstream tables.
Descompone un objetivo de procesamiento en un flujo de nodos Chains (source → transform → sink). Use when the user wants to build a data pipeline / chain, or describes an end-to-end "load X, clean it, summarize, export" goal.
| name | EDA — Exploración Inicial |
| description | First look at a dataset — profiles structure, data quality, and key distributions to build a mental model before any specific analysis. Use when the user wants to understand a new table or asks for an overview without a defined goal. |
| keywords | explore, eda, overview, analyze, profile, summarize, dataset, explorar, analizar, resumen, exploración, datos, qué hay, entender, primer, vistazo, examinar, describir, overview |
| next | data-storytelling |
Activa cuando el usuario quiere entender un dataset por primera vez, sin un análisis específico definido. Objetivo: formar rápido un modelo mental —estructura, calidad y lo que vale la pena investigar después— en un solo pase eficiente.
Esto es un workflow (hay un orden sensato), pero las queries que ejecutes son decisión de juicio: deja que lo que descubras guíe la siguiente pregunta, no corras queries por reflejo.
attach_file con la ruta exacta. Si no, read_file con mode='list' para descubrir los disponibles.list_tables (nombres exactos) → describe_table (tipos, columnas, muestra).profile_data: null %, únicos, rangos, top values. Reporta problemas de calidad primero.display_chart, eligiendo el tipo con el marco de "Chart Selection" / storytelling (NO por defecto line solo porque haya una fecha; con pocos períodos es comparación → barras).final_answer con: hallazgos de calidad + 3 insights clave + sugerencia del siguiente análisis.caveats de final_answer.time-series, metric-investigation o data-quality.