Code Demonstration

Create working code demonstrations and examples.

Purpose

Generate demonstrations for:

• Package vignettes
• Tutorial materials
• Conference presentations
• Documentation examples
• Teaching code

Required Input

• Topic: What to demonstrate
• Audience: Beginners, intermediate, experts
• Format: Vignette, slides, standalone script
• Length: Brief example or comprehensive tutorial

Process

1. Define scope

   • What concepts to cover
   • Prerequisite knowledge assumed
   • Learning objectives

2. Design flow

   • Build from simple to complex
   • Include intermediate outputs
   • Show common variations

3. Write code

   • Clear, readable style
   • Helpful comments
   • Runnable examples

4. Add context

   • Explain the "why"
   • Note common pitfalls
   • Suggest next steps

Output Format

# Demo: [Topic]

## Overview
[What this demo covers and why it matters]

## Prerequisites
- [Required packages]
- [Assumed knowledge]

## Setup
```r
# Load packages
library(package1)
library(package2)

# Sample data
data <- create_example_data()

Part 1: [First Concept]

[Explanation of what we're doing]

# Step 1: Description
result <- function_call(data)
result
#> [Expected output shown]

[Interpretation and notes]

Part 2: [Second Concept]

[Continue building...]

Common Variations

Variation A: [Description]

# How to modify for this case

Variation B: [Description]

# Alternative approach

Troubleshooting

Issue: [Common problem] Solution: [How to fix]

Next Steps

• [Where to learn more]
• [Related functions/topics]

References

• [Documentation links]
• [Papers if applicable]

## Demo Types

### Quick Example
- Single concept
- ~10 lines of code
- Immediate output

### Tutorial
- Multiple concepts
- Step-by-step progression
- Full workflow

### Vignette
- Comprehensive coverage
- Real-world use case
- Publication quality

### Live Coding
- Presentation format
- Clear stopping points
- Audience interaction notes

## MCP Integration

Uses these tools when available:
- `r_execute` - Run and verify code
- `r_plot` - Generate visualizations
- `r_preview` - Preview output

## Examples

**Request**: "Create a demo of linear regression in R for beginners"

**Request**: "Generate a vignette showing mediation analysis workflow"

**Request**: "Demo ggplot2 customization for publication-quality figures"

**Request**: "Create live-coding examples for a dplyr workshop"

## Best Practices

### Code Style
- Use consistent formatting
- Include all library() calls
- Show expected output in comments

### Explanation
- Explain before code blocks
- Interpret after outputs
- Connect to real-world use

### Reproducibility
- Set seeds for random operations
- Include data creation code
- Note package versions

## Tips

- Start with the simplest working example
- Build complexity gradually
- Include "what if" variations
- Show both success and error cases
- Link to further resources