Profile datasets to generate data dictionaries, statistics, and observations.
npx skills add https://github.com/thbraet/claude-template --skill describe-data
Please help me install this Agent Skill. Skill: describe-data Source: https://github.com/thbraet/claude-template/tree/main/skills/describe-data Command: npx skills add https://github.com/thbraet/claude-template --skill describe-data
DescribeData profiles each acquired dataset to produce a comprehensive data dictionary, surface statistics, and initial observations, enabling faster data understanding and preparation.
Profile a dataset located at data/raw/sales.csv to generate the 2.2 data description artifacts.
High-intent search queries and answers about installing and using this skill.
To generate a data dictionary, you profile the dataset to map field names, types, descriptions, and surface statistics. This produces a Jupyter notebook and a Markdown summary for downstream exploration.
Surface statistics are generated by profiling datasets to calculate row counts, nulls, unique values, and value distributions. This process also produces initial observations to accelerate data understanding.
Data profiling integrates with CRISP-DM 2.2 by targeting datasets listed in the 2.1 data collection report. It generates artifacts like a structured Markdown summary and notebook for the 2.2 data description phase.
Yes, dataset profiling automatically maps each field to a data dictionary entry, surface statistics, and modeling roles. This eliminates manual coding by generating the documentation directly from the acquired data.
Dataset profiling outputs a Jupyter notebook and a structured Markdown summary document. These artifacts contain the generated data dictionary, surface statistics, and initial observations for downstream use.
No specific external dependencies are required to execute the Python-based profiling. The skill runs standalone to generate the data dictionary, surface statistics, and modeling role artifacts.
thbraetDocument data acquisition and initial inventories for CRISP-DM workflows.
agentic-inRun interactive Jupyter notebook sessions with live kernel management.
eren726290Apply descriptive statistics, trend analysis, outlier detection, and hypothesis testing to data using Python.
thbraetTranslate business objectives into data mining goals with success criteria.
