skilld
All skills
oaustegard avatar

/exploring-data

@fde9b75
by Oskar Austegardoaustegard/claude-skills150 stars
6

Exploratory data analysis. Use when users upload .csv/.xlsx/.json/.parquet files or request "explore data", "analyze dataset", "EDA", "profile data". Small files get ydata-profiling HTML/JSON reports; large files (over 200MB or 5M rows) get fixed-memory DuckDB/sketch profiling. Also covers near-duplicate row detection, cross-file key overlap ("can these join?"), dataset drift vs a stored baseline, and time-series profiling.

Ask your Agent

Use this Skill: https://skilld.dev/gh/oaustegard/claude-skills/exploring-data

This session only. Nothing lands on disk.

referencesUSAGE.md

≈460 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Exploring Data - Usage Documentation

What This Skill Does

Generates comprehensive exploratory data analysis reports using ydata-profiling, a battle-tested library with 12.5k+ GitHub stars.

Features

Minimal Mode (default, 5-10s):

  • Dataset overview (rows, columns, types, memory)
  • Variable analysis (distributions, statistics)
  • Missing value analysis
  • Duplicate detection
  • Correlation matrices
  • Data quality alerts

Full Mode (10-20s):

  • Everything in minimal +
  • Scatter plot matrices
  • Sample data display
  • Character/script analysis (for text)
  • Auto-correlation
  • More sophisticated visualizations

Output Formats

HTML (default):

  • Interactive, self-contained report (500KB-2MB)
  • Tabbed navigation
  • Embedded visualizations
  • Downloadable via browser

JSON:

  • Machine-readable format
  • For programmatic access
  • Can be parsed by other tools

Installation

First use only (~19s):

bash /mnt/skills/user/exploring-data/scripts/install_ydata.sh

Creates isolated venv at /home/claude/.venvs/exploring-data with:

  • ydata-profiling 4.17.0
  • ~50 dependencies
  • Total: 559MB

Subsequent uses: Instant (venv already exists)

Performance

Dataset Size Minimal Mode Full Mode
<10k rows 2-5s 5-10s
10-50k rows 5-10s 10-15s
50-100k rows 10-15s 15-25s

Troubleshooting

If installation fails:

  • Check network access to pypi.org
  • Ensure uv is available: uv --version
  • Try manual install:
    uv venv /home/claude/.venvs/exploring-data
uv pip install ydata-profiling setuptools --python /home/claude/.venvs/exploring-data

If analysis fails:

  • Check file format is supported (.csv, .xlsx, .json, .parquet, .tsv)
  • Verify file exists at specified path
  • Check file is not corrupted

Source: SKILL.md on GitHub

Third-party checks

3 warnings13d4 checks · Risk SAFE
  • Gen Agent Trust Hub13d

    The skill provides robust exploratory data analysis capabilities using industry-standard libraries like ydata-profiling and DuckDB. It manages dependencies securely via virtual environments and includes logic for handling both small and large datasets. All external resources originate from trusted repositories and well-known services. No malicious patterns were detected.

  • Socket13d

    1 alert: gptAnomaly

  • Snyk13d

    Risk: MEDIUM · 1 issue

  • Runlayer7mo

    11/11 files flagged

Provenance

Signed by skilld at fde9b75. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 4 weeks ago

Capability

metadata
{
  "version": "0.2.0"
}

README badge

README badge for oaustegard/claude-skills/exploring-data