A professional library of reusable prompt patterns and structural frameworks for high-fidelity AI image generation across multiple models.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install image-prompt-patterns
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install image-prompt-patterns using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Image Prompt Patterns?
Image Prompt Patterns is a specialized knowledge-layer skill designed to bridge the gap between creative intent and high-quality AI visual output. By codifying prompt engineering into five distinct architectural styles—Photographic, Cinematic, Structured JSON, Cultural Mashup, and Narrative—this skill ensures precision when working with models like Midjourney, Nano Banana Pro, and Flux.
Unlike simple generators, this skill focuses on the logical structure of a prompt, incorporating layered stacking, controlled contradictions, and named light sources. It serves as an essential repository for Openclaw Skills users who need to translate complex concepts into specific visual descriptors without relying on trial and error.
Image Prompt Patterns Use Cases
- Creating professional commercial photography and social media mockups using 8-layer structural prompting.
- Developing stylized cinematic storyboards and anime-inspired battle scenes with dynamic composition.
- Generating high-density information graphics and UI/UX landing page mockups using structured JSON schemas.
- Migrating existing prompt styles between different AI models (e.g., MJ to GPT-Image-2) while maintaining visual consistency.
- Creating character sheets and multi-view reference images for consistent character design.
How Image Prompt Patterns Works
- Analyze user intent to determine the required visual output type (e.g., portrait, UI, or narrative scene).
- Select the optimal prompt architecture from the five core frameworks (Type A through E).
- Apply specialized techniques like Layered Stacking for anatomical detail or Named Light Sources for cinematic depth.
- Parameterize the prompt using bracketed arguments to allow for easy user iteration and customization.
- Refine and optimize the generated text based on specific model strengths (e.g., natural language for GPT-Image-2 vs. tag-stacking for MJ).
- Permanently archive successful custom logic and 'lessons learned' into the reference repository for continuous improvement.
Image Prompt Patterns Setup
This is a knowledge-based skill that operates within the AI agent's logic layer. To implement its patterns in your workflow, ensure your environment has access to the reference library.
# Standard usage involves querying the skill's internal reference files
cat ./skills/image-prompt-patterns/references/portrait.md
Note: This skill provides the prompt text; actual image execution requires a separate execution tool like gencut-image.
Image Prompt Patterns Data Schema & Taxonomy
The skill organizes its patterns into a taxonomy based on visual intent and technical structure:
| Category | File Path | Focus Area |
|---|---|---|
| Portraits | references/portrait.md |
Anatomical detail, skin texture, and lighting |
| UI/Web | references/ui-mockup.md |
Layout, components, and interface mockups |
| Structured | references/structured-json-prompt.md |
Complex, info-dense data visualizations |
| Cinema | references/cinematic-storyboard.md |
Artistic style, camera angles, and atmosphere |
| Templates | references/parametric-template.md |
Reusable variables and cross-style logic |
Image Prompt Patterns Advanced Features
- Structured JSON Prompting: Uses pseudo-code to define precise spatial layouts for infographics and landing pages.
- Cultural-Fusion Mashups: Leverages native language 'style anchors' to blend distinct cultural aesthetics (e.g., Irasutoya + Cyberpunk).
- Parametric Template Mechanism: Allows for bulk production of different scenes featuring the same character using {argument} injection.
- Auto-Refinement Loop: A mandatory contribution rule ensures the skill 'learns' from every complex prompt generation by updating the reference library.
SKILL.md
Loading
Related Openclaw Skills
TradeAlpha News V2
TradeAlpha News V2 provides AI agents with real-time global financial news and high-precision semantic search capabilities for market analysis.
CC Ride-Hailing Passenger Prototype
A specialized design system skill for extending the CC Ride-Hailing passenger-side HTML/CSS prototypes with strict architectural consistency.
12306 Train Query Skill
An AI-ready tool for retrieving real-time train schedules and seat availability directly from China's 12306 railway system.
Backlink Strategy
A comprehensive framework for ecommerce stores to build domain authority through high-value outreach, partnership identification, and the creation of linkable content assets.
SEO Blog Planner
A strategic content planning tool that transforms ecommerce blogs into high-conversion sales funnels through keyword-driven topic clusters.
Travel Disruption Rescue
A structured emergency response workflow designed to transform travel chaos into an actionable recovery plan for cancellations, delays, and lost baggage.
Featured*
