Exploratory Data Analysis (Safe Mode)

A fail-closed EDA skill that profiles authorized local scientific files, audits missingness, leakage and outliers, and drafts a rigorous report without exposing raw values.

Data & AnalyticsAdvanced★ 33,030⑂ 3,248AI score 9/10Last updated: Aug 9, 2026

What it does

Produces bounded, deterministic schema and profile reports for CSV/TSV/JSON using only the Python standard library.
Audits missingness patterns, group/entity/time-based split overlap, and common leakage traps before modeling.
Compares mean/SD against median/IQR/MAD and shows outlier and log-transformation sensitivity — it never deletes, imputes, or normalizes data for you.
Optionally inspects metadata only for NumPy (.npy/.npz), HDF5, FASTA/FASTQ, and PNG/JPEG/TIFF/OME-TIFF (no pixel decoding, no pickle, no network).
Generates a Markdown report scaffold for assumptions, sensitivity analyses, and limitations.
Unlisted formats fail closed; all file-derived text (headers, cells, tags, attributes) is treated as untrusted data and never executed.

Who it's for

Researchers handling sensitive clinical, omics, or imaging data who need first-pass exploration without leaking identifiers.
Data scientists who must verify split boundaries and leakage risk before training.
Teams that need reproducible, auditable EDA with recorded commands and versions.

Examples

Run python scripts/capability_manifest.py inspect data.csv --root /approved/project to confirm support, then profile the file.
Use missingness_leakage_audit.py --entity-column subject_id --split-column split to catch the same subject appearing in both train and test.
Use distribution_sensitivity.py --column measurement to see how outliers move the mean and whether a log transform is warranted.
Use report_scaffold.py to produce a tokenized, identifier-free EDA report draft.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/HEAD/skills/exploratory-data-analysis/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)

Download the ZIP with the button below.
In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.

↓ Download ZIP

Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/exploratory-data-analysis folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/exploratory-data-analysis/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/exploratory-data-analysis ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

Open a terminal and go to your working directory.
Clone the repository: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
Create the skills folder: mkdir -p ~/.claude/skills
Copy this skill: cp -r scientific-agent-skills/skills/exploratory-data-analysis ~/.claude/skills/
Check Python: python3 --version (3.11+ required; 3.12+ for the optional NumPy/tifffile inspectors).
Install optional format support only if needed: uv pip install "numpy==2.5.1" "h5py==3.16.0" "biopython==1.87" "pillow==12.3.0" "tifffile==2026.7.14"
Set up an approved data root (e.g. /approved/project) containing only copies you are allowed to analyze.
Restart Claude Code and ask something like "Profile this CSV and audit it for missingness and leakage."

View source on GitHub ↗License: MIT