exploratory-data-analysis

Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain formats are reference-only and unknown formats fail closed.

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

92%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is action-dense, well-structured, and operationally rigorous with validation checkpoints and one-level-deep references that all exist on disk. The only minor drag is dated version-pinning content that is not isolated in a deprecated section.

Suggestions

Move the dated version-baseline table and pinned install commands into a dedicated 'Pinned dependency snapshot' or 'old patterns' section so the time-sensitive pins don't dilute the evergreen core guidance.

Consider condensing the seven-row package version table to the minimal pin set plus a pointer to a reference file for the full dated snapshot, reducing token cost for the common no-optional-deps case.

Dimension	Reasoning	Score
Conciseness	The body is dense and lean with no padding of concepts Claude already knows, but the dated version-baseline table with per-package published dates is time-sensitive content outside a deprecated/old-patterns section, which the rubric says should lightly penalize conciseness.	4 / 5
Actionability	Provides fully executable, copy-paste-ready commands with concrete flags (--root, --output, --max-rows, column/group/entity/split/time flags) for every common case, plus a runnable capability registry, matching the top anchor.	5 / 5
Workflow Clarity	A five-step workflow has explicit validation checkpoints — manifest-before-content with 'If status is reference_only, do not run' and 'If unknown, stop', authorization/root confirmation with stop-and-ask feedback — fitting the anchor with feedback loops for risky operations.	5 / 5
Progressive Disclosure	A clear overview points one level deep to six real, verified reference files (each with a scope table), bundled scripts, and a report-template asset; content is appropriately split with easy navigation and no nested-reference chains.	5 / 5
	Total	19 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, trigger-rich, and answers both what and when with concrete file-format keywords, plus a clear fail-closed boundary that distinguishes it from generic data skills. No meaningful gaps relative to the rubric anchors.

Dimension	Reasoning	Score
Specificity	Names the domain and lists multiple concrete actions — 'redacted CSV/TSV/JSON profiles', 'image metadata inspection', 'missingness/leakage audits', 'outlier and transformation sensitivity', 'EDA report scaffolds' — giving comprehensive coverage rather than generic verbs.	5 / 5
Completeness	Explicitly answers both what ('bounded, local exploratory analysis of explicitly supported scientific files' plus enumerated capabilities) and when ('Use for...') with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Comprehensive natural keyword and extension coverage including 'CSV/TSV/JSON', 'NumPy', 'HDF5', 'FASTA/FASTQ', 'EDA', and 'missingness/leakage', matching the kinds of phrases and file formats a user would actually name.	5 / 5
Distinctiveness Conflict Risk	The bounded-local / fail-closed framing and the explicit scientific-file scope carve a clear niche unlikely to trigger for unrelated skills.	5 / 5
	Total	20 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: K-Dense-AI/scientific-agent-skills
Path: skills/exploratory-data-analysis/SKILL.md
Commit: b2a92ba

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.