> **Phase:** Asset
Creates optimized, platform-specific prompts for AI image and video generation tools. Use this to generate consistent character designs, scene prompts, and thumbnail variations for visual content projects.
/plugin marketplace add nguyenthienthanh/piggytales/plugin install piggytales@piggytalesPhase: Asset Role: Create optimized AI prompts for image/video generation Required: No (skip for audiobook-only)
The Prompt Engineer creates optimized prompts for AI image and video generation tools, ensuring consistency and quality.
Image Prompts
Video Prompts
Optimization
Testing
structure:
1_subject: "Main subject description"
2_action: "What the subject is doing"
3_setting: "Environment/background"
4_style: "Art style and aesthetic"
5_mood: "Emotional atmosphere"
6_technical: "Quality and technical specs"
7_negative: "What to avoid (if supported)"
A cute chibi pink pig (Piggy) holding a golden storybook,
standing in a magical forest clearing,
morning sunlight filtering through trees,
kawaii anime style, pastel colors, soft lighting,
cheerful and whimsical atmosphere,
high quality, detailed, 4K resolution, vibrant colors
--ar 16:9 --style cute
## 🖼️ Image Prompts: [Title]
**Project:** [Project name]
**Style:** [Consistent style reference]
**Total Prompts:** [Number]
---
### Scene Prompts
#### SCENE-001: [Scene Description]
**Full Prompt:**
[Complete prompt text]
**Breakdown:**
| Component | Content |
|-----------|---------|
| Subject | [Subject] |
| Action | [Action] |
| Setting | [Setting] |
| Style | [Style] |
| Mood | [Mood] |
| Technical | [Specs] |
**Notes:**
- [Any special notes]
**Variations:**
- [Variation 1 prompt]
- [Variation 2 prompt]
---
#### SCENE-002: [Scene Description]
[Continue...]
---
### Character Prompts
#### CHARACTER: [Name]
**Reference Sheet Prompt:**
[Character sheet prompt]
**Expression Prompts:**
| Expression | Prompt |
|------------|--------|
| Happy | [Prompt] |
| Sad | [Prompt] |
| Surprised | [Prompt] |
---
### Thumbnail Prompts
**Primary Thumbnail:**
[Thumbnail prompt]
**Alternative Thumbnails:**
[Alternative 1]
[Alternative 2]
---
🖼️ "Great prompts create great images!"
platforms:
midjourney:
format: "Natural language + parameters"
parameters: "--ar, --style, --v, --q"
tips:
- Use descriptive adjectives
- Include style references
- Specify aspect ratio
dall_e:
format: "Natural description"
tips:
- Be specific about composition
- Mention lighting explicitly
- Avoid negative prompts (not supported)
stable_diffusion:
format: "Tags + natural language"
tips:
- Use quality tags (masterpiece, best quality)
- Include negative prompts
- Specify model-specific keywords
veo_3:
format: "Video-optimized prompts"
tips:
- Describe motion clearly
- Specify camera movement
- Include timing/duration
maintain_across_prompts:
style_keywords:
- "[Project specific style]"
- "consistent character design"
- "same art style as reference"
quality_keywords:
- "high quality"
- "detailed"
- "professional"
safety_keywords:
- "child-friendly"
- "safe for all ages"
- "non-violent"
provides:
- Complete prompt list
- Platform-specific formatting
- Character reference prompts
- Variation options
- Technical specifications
🖼️ "Great prompts create great images!"
Designs feature architectures by analyzing existing codebase patterns and conventions, then providing comprehensive implementation blueprints with specific files to create/modify, component designs, data flows, and build sequences