review-process

Audit CRISP-DM phase 1-4 artifacts and generate a draft 5.2 review report and notebook.

Updated Mar 20, 2026

Search Tags:#documentation #evaluation #data-science #model-governance #artifact-audit #crisp-dm #process-review

Ready to run

npx skills add https://github.com/thbraet/claude-template --skill review-process-thbraet

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review-process
Source: https://github.com/thbraet/claude-template/tree/main/skills/review-process
Command: npx skills add https://github.com/thbraet/claude-template --skill review-process-thbraet

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables structured retrospectives of CRISP-DM engagements, surfacing what worked, what didn’t, and where methodology gaps exist to guide deployment decisions and continuous improvement.

Core Features & Use Cases

Assess phase-by-phase execution quality across CRISP-DM phases 1–4 plus the evaluation outputs
Conduct cross-phase consistency checks to ensure alignment between business objectives, data understanding, preparation, modeling, and evaluation
Identify overlooked factors and generate actionable lessons learned to inform future projects
Produce artifact outputs: a formal 5.2 review document and an accompanying notebook for programmatic audits and dashboards
Provide a structured health verdict and recommended actions before deployment

Quick Start

Run the review-process skill to generate a 5.2 CRISP-DM review by auditing phase artifacts and producing a structured report and accompanying notebook.

Frequently Asked Questions about review-process

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I conduct a CRISP-DM phase review for data science project artifacts?▼

To conduct a CRISP-DM phase review, audit project artifacts from phases 1 through 4 to surface process-quality insights, cross-phase consistency, and overlooked factors. This generates a structured health verdict and a formal 5.2 review document to guide deployment decisions.

What is the best way to check cross-phase consistency in a CRISP-DM data science engagement?▼

Checking cross-phase consistency involves auditing alignment between business objectives, data understanding, preparation, modeling, and evaluation. The review process evaluates these programmatic audits to ensure methodology gaps are identified before final deployment.

How do I generate a model governance artifact audit report for production readiness?▼

Generating a model governance artifact audit produces a formal 5.2 review document at docs/crisp-dm/5-evaluation/5.2-review-process.md and an accompanying notebook. It evaluates execution quality and outputs recommended actions for production readiness.

When do I need a process review for CRISP-DM data science projects?▼

A process review is needed during phase 1 through 4 evaluations to conduct structured retrospectives. It identifies overlooked factors, generates actionable lessons learned, and provides a structured health verdict before making final deployment decisions.

Can I programmatically audit CRISP-DM documentation to find methodology gaps?▼

Yes, you can programmatically audit CRISP-DM documentation using a generated notebook at notebooks/5.2-review-process.ipynb. It integrates with existing docs to identify overlooked factors and surface methodology gaps across data science engagements.