# prompts
Show GenAI prompt templates for sprite generation.
## Usage
```
pxl prompts [TEMPLATE]
```
## Arguments
| `[TEMPLATE]` | Template name to show. If omitted, lists all available templates |
## Available Templates
| `character` | Character sprite generation prompt |
| `item` | Item and object sprite generation |
| `tileset` | Tileset and terrain generation |
| `animation` | Animation sequence generation |
## Description
The `prompts` command provides ready-to-use prompt templates for generating Pixelsrc content with AI models. Each template includes:
- Clear instructions for the AI
- Format specifications
- Example output structure
- Common parameters to customize
## Examples
### List available templates
```bash
# Show all available templates
pxl prompts
```
### Get specific template
```bash
# Get character sprite template
pxl prompts character
# Get animation template
pxl prompts animation
```
### Use with AI tools
```bash
# Copy template to clipboard (macOS)
# Pipe to AI tool
# Save template for editing
pxl prompts tileset > tileset-prompt.txt
```
### Preview AI output quickly
```bash
# After generating sprite, preview in terminal
pxl show sprite.pxl -s character
# View with coordinates for debugging
pxl grid sprite.pxl -s character
# Get structural breakdown
pxl explain sprite.pxl -s character
```
## Sample Template Output
```
# Character Sprite Generation
Generate a pixel art character sprite in Pixelsrc format.
## Requirements
- Size: 16x16 pixels
- Style: Retro game aesthetic
- Include: Front-facing idle pose
## Output Format
Create a palette with 4-8 colors, then a sprite referencing those colors.
Example structure:
```
palette:
name: character_colors
colors:
outline: #000000
skin: #E0A070
...
sprite:
name: character
palette: character_colors
grid:
_ _ _ _ ...
```
## Guidelines
- Use `_` for transparent pixels
- Keep outlines consistent (usually 1px black)
- Center the character in the grid
...
```
## Use Cases
- **Quick start**: Get working prompts without writing from scratch
- **Consistency**: Ensure AI outputs match expected format
- **Training**: Understand effective prompt structure
- **Integration**: Build into AI-powered workflows
## See Also
- [prime](prime.md) - Format specification for context injection
- [validate](validate.md) - Validate AI-generated output
- [suggest](suggest.md) - Fix common AI generation errors