data-science-python

Python data science: notebook structure, data validation, reproducibility, and model documentation

1.11x

Quality

60%

Does it follow best practices?

Impact

76%

1.11x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Fix and improve this skill with Tessl

tessl review fix ./skills/data-science-python/SKILL.md

Quality

Content

87%

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

This is a well-crafted data science skill that is concise, actionable, and well-organized. Its main strength is providing executable code examples alongside practical conventions without over-explaining concepts Claude already knows. The only notable weakness is the lack of explicit validation checkpoints and error recovery feedback loops between workflow stages, particularly around feature engineering and model evaluation steps.

Suggestions

Add explicit validation/feedback loops between workflow stages, e.g., 'After feature engineering, validate: assert no data leakage between train/test splits' and 'If evaluation metrics are below threshold, revisit feature engineering before proceeding.'

Dimension	Reasoning	Score
Conciseness	The content is lean and efficient. It doesn't explain what notebooks are, what pandas does, or other concepts Claude already knows. Every section delivers actionable information without padding. The model documentation section is a concise checklist rather than verbose explanation.	3 / 3
Actionability	Provides fully executable code examples for data validation (assert statements and pandera), reproducibility (seed setting), and sklearn pipelines. Concrete naming conventions (model_rf_20260115_v1.pkl) and specific tool recommendations (MLflow, pandera) make guidance immediately actionable.	3 / 3
Workflow Clarity	The notebook structure provides a clear sequence, and data validation includes a validation step after loading. However, the overall data science workflow lacks explicit validation checkpoints between stages (e.g., validate after feature engineering, validate model outputs before saving). There's no feedback loop for error recovery when validation fails.	2 / 3
Progressive Disclosure	For a skill under 50 lines with no need for external references, the content is well-organized into clearly labeled sections that serve as a concise overview. Each section is appropriately scoped and the structure supports easy navigation without requiring separate files.	3 / 3
	Total	11 / 12 Passed

Description

32%

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 reads as a list of topic areas rather than a functional skill description. It lacks action verbs describing what the skill does, has no 'Use when...' clause to guide selection, and could benefit from more specific trigger terms like 'Jupyter', 'pandas', or '.ipynb'. The noun-phrase style makes it unclear whether this skill creates notebooks, reviews them, or provides best practices.

Suggestions

Add explicit action verbs describing what the skill does, e.g., 'Reviews and structures Jupyter notebooks for reproducibility, validates data pipelines, and documents ML models.'

Add a 'Use when...' clause with trigger terms, e.g., 'Use when the user asks about Jupyter notebooks, data science workflows, pandas pipelines, sklearn models, .ipynb files, or reproducible analysis.'

Include common user-facing terms like 'Jupyter', 'pandas', 'machine learning', 'DataFrame', '.ipynb' to improve trigger term coverage and distinctiveness.

Dimension	Reasoning	Score
Specificity	Names the domain (Python data science) and some areas (notebook structure, data validation, reproducibility, model documentation), but these are more like topic categories than concrete actions. No verbs describing what the skill actually does.	2 / 3
Completeness	Partially addresses 'what' through topic listing but lacks any explicit 'when' clause or trigger guidance. Per rubric guidelines, a missing 'Use when...' clause caps completeness at 2, and the 'what' is also weak (topics rather than actions), warranting a score of 1.	1 / 3
Trigger Term Quality	Includes relevant keywords like 'Python', 'data science', 'notebook', 'data validation', 'reproducibility', and 'model documentation', but misses common variations users might say like 'Jupyter', 'pandas', 'sklearn', 'machine learning', '.ipynb', or 'data analysis'.	2 / 3
Distinctiveness Conflict Risk	The combination of Python data science with notebook structure and reproducibility provides some specificity, but could overlap with general Python coding skills, data analysis skills, or documentation skills. The scope is broad enough to create potential conflicts.	2 / 3
	Total	7 / 12 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 — 11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository: ucdavis/ai-skills-registry
Commit: c0b2e4b

Reviewed: 23 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.