🎬 From Simple Ideas to Hollywood Magic: Introducing the WAN 2.2 Prompt Enhancer

Ever tried generating AI video content and gotten... well, let's just say "amateur home video" results when you were dreaming of something more cinematic? You're not alone. The difference between a basic prompt and a professional one can mean the difference between a wobbling phone camera recording and a Hollywood blockbuster scene.

Today, I'm excited to open-source a tool that bridges this gap: the WAN 2.2 Prompt Enhancer – a Python CLI that transforms simple ideas into professional, cinematic video generation prompts using Google's Gemini AI and the comprehensive WAN 2.2 aesthetic control system.

The Problem: Great Ideas, Basic Execution

AI video generation tools like RunwayML, Pika, and others are incredibly powerful, but they're only as good as the prompts you give them. Most of us start with something like:

"a dragon flying over mountains"

And while that's a perfectly good idea, it's missing all the technical cinematography elements that make video content truly compelling. Professional video creators know to specify lighting conditions, camera angles, shot composition, color grading, movement dynamics, and stylistic approaches.

But who has time to become a cinematography expert just to generate better AI videos?

The Solution: AI-Powered Prompt Engineering

The WAN 2.2 Prompt Enhancer solves this by leveraging two powerful technologies:

1. Google's Gemini AI - For intelligent prompt transformation
2. WAN 2.2's Advanced Formula - A comprehensive aesthetic control system that includes:
3. Subject + Scene + Motion + Aesthetic Control + Stylization
4. Professional lighting terminology
5. Camera movement specifications
6. Color grading and composition rules

Here's what happens when you input that simple dragon prompt:

Input: bash $ uv run wan_prompt_enhancer.py "a dragon flying over mountains"

Actual Output:

Reading inline script metadata from `wan_prompt_enhancer.py`
🎯 Enhancing: 'a dragon flying over mountains'
✨ Generating cinematic variations...
============================================================
1.  An epic, high-fantasy dragon, scales glinting with a deep sapphire hue. It soars with powerful, rhythmic wingbeats over a jagged, snow-capped mountain range under a high noon sun. The lighting is direct and hard, emphasizing sharp relief and deep shadows across the craggy peaks. A majestic, wide-angle establishing shot, with the dragon positioned against a vast, azure sky, uses the rule of thirds for grandeur. A 24mm wide-angle lens captures the expansive landscape. The color tone is cool and desaturated, highlighting the stark beauty of the alpine environment. The camera slowly tracks the dragon's trajectory, emphasizing its effortless glide.

2.  A mystical, ancient dragon, its form wreathed in subtle, iridescent mist, ascends majestically from a cloud-filled chasm. The scene is lit by an otherworldly, diffused backlighting, emanating from a hidden aurora borealis, casting a soft, ethereal glow. The time of day is twilight, with a dramatic, medium close-up, low-angle shot, looking up at the dragon as it breaks through the cloud cover. The composition centers the dragon's powerful head and wings. A 50mm prime lens captures the intricate details of its leathery wings and glowing eyes. The color tone is a vibrant, magical palette of deep purples, greens, and blues, with soft light. The camera slowly cranes up, following the dragon's ascent, its body twisting gracefully.

3.  A menacing, battle-scarred dragon, its obsidian scales reflecting the fiery glow of an approaching storm, glides ominously above a volcanic mountain range. The lighting is dramatic and contrasty, with flashes of lightning illuminating the scene from above, creating brief, stark highlights and deep, threatening shadows. Overcast and stormy sky. A dynamic, medium shot, handheld style, focuses on the dragon's intense gaze and powerful, leathery wings. A 35mm lens emphasizes its imposing presence against the turbulent sky. The color tone is a muted, desaturated palette of grays, deep reds, and oranges, with harsh, directional light. The camera uses a quick, sweeping lateral track, conveying the dragon's swift and powerful movement, as it banks sharply around a peak.

4.  A graceful, elegant dragon, its skin shimmering with a pearlescent sheen, performs an intricate aerial ballet amidst a serene, sun-drenched mountain valley. Golden hour sunlight bathes the scene in warm, soft, directional light from the side, highlighting the dragon's delicate wings and the lush, green foliage below. A breathtaking, extreme wide shot, utilizing an aerial drone perspective, captures the dragon's intricate loops and dives against the backdrop of rolling hills and distant peaks. A 20mm wide-angle lens emphasizes the vastness and beauty of the landscape. The color tone is a warm, golden, and verdant palette, with soft, natural light. The camera performs a smooth, sweeping orbital movement around the dragon, capturing its fluid, balletic flight from multiple angles.

5.  A colossal, ancient dragon, its silhouette defined against the radiant glow of a full moon, slowly circles a colossal, snow-dusted peak under a clear, crisp night sky. The lighting is low-key and dramatic, with the moon acting as the primary, hard back light source, creating deep, atmospheric shadows. A cinematic, long shot, employing a high-angle perspective, showcases the dragon's immense scale relative to the mountain. A 85mm prime lens compresses the perspective, emphasizing the dragon's isolation. The color tone is a cool, monochromatic blue and silver, with the deep shadows creating a sense of mystery and grandeur. The camera remains static, observing the dragon's slow, deliberate circling motion, emphasizing its ancient presence.
============================================================
✅ Enhancement complete! Use these prompts in your video generation tool.

Notice how each variation gives you a completely different cinematic approach:

• Variation 1: Epic high-fantasy with hard noon lighting and wide-angle majesty
• Variation 2: Mystical twilight with aurora backlighting and ethereal mist
• Variation 3: Menacing storm scene with dramatic lightning and handheld camera work
• Variation 4: Graceful golden hour ballet with aerial drone perspectives
• Variation 5: Ancient moonlit encounter with monochromatic blues and static observation

Each maintains your core concept (dragon + mountains) while offering completely different moods, technical approaches, and visual storytelling opportunities.

Key Features That Make This Special

🎯 Professional Cinematography Knowledge Built-In

The tool understands the WAN 2.2 aesthetic control system, including:

- Lighting Types: Soft lighting, hard lighting, edge lighting, practical lighting, and more

- Camera Movements: Dolly shots, tracking shots, arc shots, compound moves

- Shot Compositions: Rule of thirds, symmetrical, left-weighted, balanced compositions

- Color Grading: Warm/cool palettes, saturated/desaturated approaches, specific color timing

⚡ Lightning Fast Setup

Getting started takes literally 2 steps:

1. Get a Google AI API Key (free at Google AI Studio)
2. Run the tool with uv (no installation needed):

Download and run in one command
uv run --from https://github.com/BitsofJeremy/WAN-2.2-Prompt-Enhancer wan_prompt_enhancer.py "your prompt here" ```

🔄 Interactive Mode for Creative Flow

Perfect for brainstorming sessions:

$ uv run wan_prompt_enhancer.py --interactive
🎬 WAN 2.2 Prompt Enhancer - Interactive Mode 💡 Tips: Be descriptive but concise
📝 Enter your basic prompt: magical forest 🎯 Enhancing: 'magical forest' ✨ Generating cinematic variations... ============================================================ 1. Fantasy Adventure Style, dappled sunlight filtering through ancient trees, medium tracking shot following a winding moss-covered path... 2. Mystical Atmosphere, ethereal mist, low-angle shot revealing towering oaks... [Plus more variations] ============================================================ ✅ Enhanced prompt #1. Ready for your next prompt!
📝 Enter your basic prompt: _ 

🎨 Multiple Stylistic Approaches

Each prompt generates 3-5 variations with different:

- Visual Styles: Photorealistic, anime, claymation, oil painting, pixel art

- Lighting Conditions: Golden hour, blue hour, overcast, neon-lit, candlelit

- Camera Work: Handheld, steadicam, drone shots, macro lenses

- Emotional Tones: Epic, intimate, mysterious, energetic, melancholic

Real-World Examples

Let me show you how this transforms various types of prompts:

Simple Character Action

Input: robot dancing in the rain

Enhanced Output Style:

Cyberpunk Street Performance, neon lighting reflecting off wet asphalt, medium close-up shot with shallow depth of field. A sleek humanoid robot with chrome plating executes fluid breakdancing moves in heavy rainfall, each movement creating cascading water droplets that catch the purple and blue neon signs...

Nature Scenes

Input: storm over ocean

Enhanced Output Style:

Epic Natural Drama, overcast lighting with dramatic storm clouds, wide-angle shot from elevated coastal cliff. Massive dark thunderheads roll across a churning ocean surface, lightning strikes illuminate towering waves crashing against jagged rocks below. The camera slowly pulls back to reveal the full scope of nature's power...

Abstract Concepts

Input: time passing

Enhanced Output Style:

Time-lapse Cinematography, mixed natural and artificial lighting, extreme wide shot transitioning to close-up. A busy city intersection transforms through multiple time periods - horse carriages dissolve into vintage cars, then modern vehicles, then futuristic transports. The camera position remains fixed while architecture morphs and seasons change...

The Technical Deep Dive

For those interested in how this works under the hood, the tool implements several smart design decisions:

Comprehensive Prompt Engineering

The system prompt I've developed teaches Gemini AI the entire WAN 2.2 methodology. It includes:

- 2,000+ words of cinematography guidance

- Specific terminology from the WAN 2.2 guide

- Examples of proper prompt structure

- Rules for generating multiple variations

Smart API Integration

Built with Google's latest genai library, featuring:

- Automatic retry logic

- Environment variable configuration

- Multiple model support (Gemini 2.5 Flash, Pro, etc.)

- Optional thinking budgets for more deliberate responses

Developer-Friendly Architecture

• Single-file deployment: Uses Python's PEP 723 inline script metadata
• Zero-installation: Works with uv run directly from GitHub
• Flexible configuration: Environment files, command-line args, or interactive prompts
• Professional error handling: Clear error messages and helpful guidance

Production Ready

The codebase includes:

- Comprehensive type hints
- Full docstring documentation
- Error handling and input validation
- Configurable models and parameters
- Interactive and batch processing modes

Getting Started Right Now

The beauty of this tool is how quickly you can start using it. Here are three ways to get going:

Option 1: Instant Trial (Recommended)

Set your API key and run immediately
export GEMINI_API_KEY="your-api-key" uv run https://raw.githubusercontent.com/BitsofJeremy/WAN-2.2-Prompt-Enhancer/main/wan_prompt_enhancer.py "a spaceship landing on an alien planet" 

Or if you've downloaded the script locally:

export GEMINI_API_KEY="your-api-key" uv run wan_prompt_enhancer.py "a dragon flying over mountains"
🎯 Enhancing: 'a dragon flying over mountains'
✨ Generating cinematic variations...
============================================================
[5 professional variations generated instantly]

Option 2: Local Setup

Clone and set up locally
git clone https://github.com/BitsofJeremy/WAN-2.2-Prompt-Enhancer.git cd WAN-2.2-Prompt-Enhancer cp .env-example .env
Edit .env with your API key
uv run wan_prompt_enhancer.py "your prompt" 

Option 3: Interactive Exploration

bash uv run wan_prompt_enhancer.py --interactive

Why I Built This (And Why It's Open Source)

As someone who spends a lot of time experimenting with AI tools, I kept running into the same frustration: amazing AI video generation capabilities held back by my amateur prompt writing skills. I'd have great creative ideas but struggle to communicate them in the technical language these tools understand best.

The WAN 2.2 guide provided an incredible framework, but it's a lot to digest and remember when you're in creative flow. I realized that AI could help AI – using Gemini to translate human creativity into the technical specifications that video generation models crave.

I'm open-sourcing this because I believe tools like this should be accessible to everyone. Whether you're:

- A content creator looking to improve your AI video output
- A developer building video generation workflows
- An educator teaching prompt engineering
- Just someone who loves experimenting with AI tools

This tool can help bridge the gap between creative vision and technical execution.

What's Next?

This is version 1.0, and I have several ideas for future enhancements:

• Multi-model support: Integration with other AI providers
• Style libraries: Pre-built collections for specific genres (horror, romance, documentary, etc.)
• Image-to-video prompts: Enhanced support for WAN 2.2's image-to-video formula
• Batch processing: Handle multiple prompts from files
• Web interface: For those who prefer browsers to command lines
• Integration guides: Specific workflows for popular video generation platforms

Try It Today

The complete source code, documentation, and examples are available now:

🔗 GitHub Repository: WAN-2.2-Prompt-Enhancer

📚 Quick Start Guide: Check the README for setup instructions

🎯 Live Demo: Run uv run https://raw.githubusercontent.com/BitsofJeremy/WAN-2.2-Prompt-Enhancer/main/wan_prompt_enhancer.py "your prompt" right now

I'd love to see what creative prompts you generate with this tool! Share your results, report bugs, or contribute improvements via GitHub issues and pull requests.

What simple prompt would you like to transform into cinematic gold? Try the tool and let me know.