---
description: Generate a professional Gemini Flash Image prompt from free-form description
argument-hint: <description>
allowed-tools: [Read, Write, Glob, Grep, Bash]
---

Create a professional image generation prompt based on the user description: "$ARGUMENTS"

Follow these steps:

1. **Context Discovery**: Automatically search for and gather relevant context from the project files:
   - Search for any mentioned characters in `/heroes/` directory
   - Look for locations in `/locations/` or `/cities/` directories
   - Find world information in `/worlds/` directory
   - Identify items in `/items/` directory
   - Check for creatures in `/creatures/` directory
   - Find applicable settings in `/settings/` directory

2. **Entity Analysis**: Parse the description to identify:
   - Characters mentioned (explicit or implied)
   - Locations referenced
   - Items involved
   - Creatures present
   - Scene context and actions
   - Appropriate style settings to apply

3. **Template Selection**: Choose appropriate template from `docs/prompting-guide.md`:
   - Prefer photorealistic scenes for character portraits and realistic scenes, or cinematic quality shots


4. **Prompt Generation**: Create a concise, standalone prompt that:
   - Uses narrative descriptions instead of keyword lists
   - Integrates project context as plain text descriptions (no references)
   - Includes specific character physical details from project files
   - Describes clothing, weapons, and items without referencing project links
   - Incorporates world architectural and environmental details
   - Specifies lighting, atmosphere, and mood
   - Includes technical specifications (photorealistic/cinematic quality)
   - Specifies aspect ratio and camera angle
   - Assumes AI model only has the prompt text and uploaded reference images

5. **Quality Standards**: Ensure the prompt follows documentation best practices:
   - Narrative structure (not keyword lists)
   - Hyper-specific details for critical elements
   - Professional photographic/cinematic terminology
   - Clear lighting and atmosphere descriptions
   - Appropriate technical quality specifications

6. **Output**: Save the final prompt as plain text in [entity_folder]/prompts or `/prepared_prompts/` if there no related entity directory:
   - Filename with timestamp and brief description
   - No markdown formatting - just the raw prompt text
   - Ready for direct copy-paste into Google AI Studio
   - Concise but comprehensive (2-3 paragraphs maximum)

**Important**: The output must be in English regardless of input language. Generate standalone prompts suitable for direct use in Google AI Studio with uploaded reference images. No project references or markdown formatting. Focus on photorealistic/cinematic quality for character and scene generation.