data-scientist

Expert data processing specialist with intelligent DuckDB/Polars selection for maximum performance. Always includes numpy, never uses pandas, runs everything through uv. Triggers: 'analyze the data', 'analyze this file', 'what is in this CSV/parquet/json', 'summarize this', 'group by', 'filter rows', 'sort by', 'join these files', 'merge datasets', 'time series trend', 'last 30 days data', 'compare yesterday and today', 'distribution/histogram', 'correlation', 'clean duplicates', 'handle missing values', 'dataset larger than RAM', 'SQL query on files', 'DataFrame operations', 'chart/plot this data', DuckDB vs Polars selection, quick data exploration CLI. NOT for plain text/code inspection, configs, or tiny inline math.

Quality

84%

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

68%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 highly actionable with executable code and well-organized references, but it carries notable redundancy (Quick Reference vs Decision Tree, Activation Context vs frontmatter) and its QA workflow lacks concrete validation checkpoints for batch data operations.

Suggestions

Remove the Quick Reference code block (it restates the Decision Tree) or merge it into the tree to cut the main redundancy.

Delete or drastically shrink the Activation Context section — those triggers already live in the frontmatter description and add ~40 lines of duplication.

Add a concrete validation step to the QA workflow, e.g. 'After execution: run `uv run scripts/quick-query.py <out> --describe` and confirm row counts/schema match expectations' so verification is executable rather than implied.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with clean code blocks and no generic concept padding, but the Quick Reference block duplicates the Decision Tree, the Activation Context section (~40 lines) restates triggers already in the frontmatter, and ABSOLUTE RULES overlaps Core Principles — more than minor trimming needed.	3 / 5
Actionability	Copy-paste-ready executable examples for DuckDB direct file queries, Polars lazy evaluation, zero-copy Arrow handoff, and the hybrid pattern, plus exact `uv run --with ...` package commands and a documented CLI — covering the common cases fully.	5 / 5
Workflow Clarity	The Quality Assurance Process (Before/During/After Execution) gives a rough sequence, but validation is implicit ('Confirm output matches expectations', 'have fallback strategy ready') with no concrete verify command; per the batch-operation cap, vague validation caps this at 3.	3 / 5
Progressive Disclosure	Clear overview with well-signaled one-level-deep references (uv-setup.md, performance-benchmarks.md, integration-patterns.md, execution-templates.md, common-scenarios.md) that are real files, but some content that could live in references (the inline Activation Context triggers, the redundant Quick Reference) is kept inline.	4 / 5
	Total	15 / 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.

This is an exemplar description: it states a concrete capability, includes natural trigger phrases with file extensions, gives explicit when-to-use guidance, and disambiguates with a negative-scope clause. Voice is appropriately third person.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('analyze the data', 'group by', 'filter rows', 'join these files', 'merge datasets', 'clean duplicates', 'handle missing values', 'chart/plot this data') plus the core capability of intelligent DuckDB/Polars selection — comprehensive coverage rather than a few actions.	5 / 5
Completeness	Explicitly answers 'what' ('Expert data processing specialist with intelligent DuckDB/Polars selection... runs everything through uv') and 'when' via a long explicit Triggers list plus a 'NOT for...' negative clause, matching the top anchor.	5 / 5
Trigger Term Quality	Comprehensive natural terms including synonyms and file extensions ('analyze the data', 'what is in this CSV/parquet/json', 'last 30 days data', 'distribution/histogram', 'correlation', 'SQL query on files') — covers the phrases a user would naturally say.	5 / 5
Distinctiveness Conflict Risk	A clear niche (high-performance DuckDB/Polars data processing via uv) with distinct triggers and a 'NOT for plain text/code inspection, configs, or tiny inline math' boundary, giving minimal conflict risk.	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: code-yeongyu/lazycodex
Path: plugins/omo/skills/data-scientist/SKILL.md
Commit: 10f9558

Reviewed: 5 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.