Visualization

Generate ArviZ diagnostic and substantive figures for Bayesian analyses.

3|Updated Jan 29, 2026

Search Tags:#visualization #diagnostics #plotting #bayesian #arviz #model-comparison #posterior

Ready to run

npx skills add https://github.com/jg-you/bayesian-workflow --skill visualization-jg-you

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: Visualization
Source: https://github.com/jg-you/bayesian-workflow/tree/main/skills/visualization
Command: npx skills add https://github.com/jg-you/bayesian-workflow --skill visualization-jg-you

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analysts often struggle to produce consistent, publication-ready figures for Bayesian model diagnostics and results, which can hinder interpretation and communication.

Core Features & Use Cases

Diagnostic Figures: ArviZ-based plots for MCMC diagnostics (trace, rank, energy, forest, posterior, PPC) using ArviZ.
Substantive Figures: visualizations that summarize posterior beliefs, effects, and model comparisons; guidelines for model-specific visuals.
Use Case: after fitting a model, generate diagnostic and substantive figures to accompany a report, ensuring clear interpretation and audience-appropriate visuals.
Asking for user input: Always ask the user what aspects of the results are most important before plotting.

Quick Start

Ask the user what figures they want, then generate and save the appropriate ArviZ plots.

Frequently Asked Questions about Visualization

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

FAQPage Schema

How do I create Bayesian diagnostic plots after fitting a model?▼

Bayesian diagnostic plots are generated using ArviZ to visualize MCMC traces, rank plots, energy, and posterior distributions. You simply specify which diagnostic figures are most important for your interpretation, and the Skill produces and saves them to your project's figures directory.

What is the best way to standardize styling for posterior summary figures?▼

Standardizing styling for posterior summary figures involves applying consistent visual guidelines during plot generation. The Skill enforces standard styling for ArviZ plots, ensuring your posterior beliefs, effects, and model comparisons are presented in a clear, publication-ready format.

Can I generate model comparison visuals using ArviZ plots?▼

Yes, ArviZ plots support model comparison visuals that summarize posterior effects and differences. The Skill generates these substantive figures after model fitting to help you communicate and interpret model comparisons effectively for your audience.

Does this approach work for both MCMC diagnostics and substantive result reporting?▼

Yes, this approach handles both MCMC diagnostics and substantive result reporting. It produces ArviZ-based diagnostic visuals like trace and energy plots alongside substantive figures that summarize posterior beliefs, ensuring comprehensive coverage for interpretation and reporting.

When do I need to generate forest plots and PPCs for Bayesian analysis?▼

Forest plots and posterior predictive checks (PPCs) are needed when you want to validate model fit and summarize posterior distributions visually. The Skill generates these specific ArviZ diagnostic figures after model fitting to support clear interpretation and audience-appropriate reporting.

Why should I ask what aspects of the results are most important before plotting?▼

Asking what aspects of the results are most important before plotting ensures the generated figures align with your interpretation goals. The Skill always requests this user input first, guaranteeing the final ArviZ visuals and diagnostics are audience-appropriate and relevant.