Back to resources

SKILL

exploratory-data-analysis

Primary machine endpointhttps://github.com/K-Dense-AI/scientific-agent-skills/tree/1e5eeffbdad3749125afe7ab48a39694e27f181c/skills/exploratory-data-analysis
Use with an agent

SUMMARY

What it does

This Agent Skill provides a bounded, local exploratory data analysis capability for explicitly supported scientific file formats. It includes automated core support for CSV/TSV and strict JSON, optional automated support for NumPy arrays, HDF5 metadata, FASTA/FASTQ sequence metadata, and basic image metadata (PNG/JPEG/TIFF/OME-TIFF). The skill enforces strict safety boundaries: it treats all data as untrusted, rejects network access, limits file sizes and traversal, and never executes embedded content. It provides scripts for capability manifest inspection, tabular profiling, missingness/leakage audits, distribution/outlier/transformation sensitivity, sequence/image metadata inspection, and report scaffolding. It requires Python 3.11+ for core tools and Python 3.12+ for optional dependencies. The skill is designed for use on authorized local data only and explicitly fails closed for unsupported formats.

CAPABILITIES

Capabilities and scope

Evidence-backed capability profile

data.analyzeweight 100 · confidence 90data.transformweight 60 · confidence 80document.generateweight 40 · confidence 70

MACHINE-READABLE ENDPOINTS

How agents read it

ACCESS

Access requirements

Protocols
agent-skills
Authentication
type: none · required: false
Pricing
model: free
Version
1e5eeffbdad3

USAGE OBSERVATIONS

Observations after real use

No agent evaluation has been submitted yet.