Core Positioning: Write AI video prompts with a director's mindset. AI is not a painter, but a photographer who just entered the industry—what you give isexecutable photographic instructions, not a description of the image.
Expression Style: Professional but not academic, give conclusions directly. Every word must be executable.
For the core execution process, see
references/core_workflow.md
Introduction
Write AI video prompts with a director's mindset—transform vague image descriptions into precise photographic instructions that AI can execute, and generate an MP4 video directly after writing.
Core Value
- Every word is executable, eliminating vague expressions, and AI accurately reproduces the envisioned image
- Generate a video with one click after writing the prompt, no need to switch tools
- Built-in diagnostic capability for 20+ common issues, directly providing fix solutions
Target Users
- 🎬 Short video creators — Professional prompts improve video quality and reduce repeated trial and error
- 📱 Self-media / content operations — Quickly produce video materials for multiple scenarios
- 🎨 Creative professionals — Transform abstract ideas into precise generation instructions
Features
Core Functions
| Function | Description |
|---|---|
| Structured prompt writing | Build prompts layer by layer according to a professional technical parameter system |
| Multimodal reference syntax | First-frame locking/style reference/camera movement reference/audio sync/subject locking |
| Golden formula verification | Check the completeness of the five elements: subject + action + scene + style + emotion |
| Common issue diagnosis | Fix issues such as character inconsistency, chaotic camera movement, and poor image quality |
| Scene template library | 8 common scene templates (natural scenery, city night scenes, character close-ups, etc.) |
| One-click video generation | Generate an MP4 video directly after writing the prompt |
Highlights
- Prompts are photographic instructions: Transform abstract descriptions into instructions that AI can execute precisely, avoiding vague expressions
- Integrated writing + shooting: After completing the prompt, you can directly generate high-definition video without switching tools
- Built-in diagnostic capability: When encountering issues such as face swapping or chaotic camera movement, it can directly provide fix solutions
Usage Guide
Authentication
Before using this skill, please ensure that the API Key has been configured.
Get API Key
- Visit Red Fox Hub official website to learn more about the service
- Go to registration page to register an account
- Newly registered users will receive free credits, and can immediately start using the API service
- After registering and logging in, obtain the API Key in the personal center, in the format
ak_xxxxxxxx
Configure API Key
REDFOX_API_KEYRetrieved from environment variables, formatak_xxxxxxxx- If not set, prompt the user to configure it themselves:
export REDFOX_API_KEY=<你的apikey>; if the user does not know how to configure it, the Agent should proactively help the user set it up:- macOS/Linux: Add
export REDFOX_API_KEY=<值>to~/.zshrc(zsh) or~/.bashrc(bash), thensourcethe corresponding file to make it take effect globally - Windows: Use
[Environment]::SetEnvironmentVariable("REDFOX_API_KEY", "<值>", "User")to set a user-level permanent environment variable (requires restarting the terminal to take effect) - After configuration is complete, verify:
echo $REDFOX_API_KEY(macOS/Linux) orecho %REDFOX_API_KEY%(Windows), to ensure that switching to another skill can also read it
- macOS/Linux: Add
Basic Usage
Directly describe the video scene you want in natural language, and the skill will automatically convert it into professional-grade prompts.
User: Help me write a video prompt for a sunset beach, with a cinematic feel
Assistant: It will build the prompt layer by layer according to technical parameters, spatial hierarchy, lighting design, and motion choreography
Common Phrasings Quick Reference
| Intent | Example Phrasing | Effect |
|---|---|---|
| Write a New Prompt | "Help me write a video prompt for a cyberpunk city night scene" | Build a complete prompt layer by layer according to a professional system |
| Effect Diagnosis | "The generated video characters always have their faces swapped, what should I do" | Diagnose the problem and provide a fix |
| Use a Template | "Give me a prompt template for product display" | Directly output a usable template for the corresponding scenario |
| Generate Video | "Use this prompt to generate a video, 16:9, 8 seconds" | Generate the video directly |
Video Generation (One-Click Output)
After writing the prompt, you can directly generate an MP4 video, supporting multiple aspect ratios and resolutions.
Supported Capabilities
| Capability | Description |
|---|---|
| Multiple Aspect Ratios | Multiple ratios available, such as landscape, portrait, and square |
| Multiple Resolutions | From standard definition to high definition, choose as needed |
| Flexible Duration | Supports 4-15 second video generation |
| Reference Image Support | You can upload a reference image to assist generation |
Generation Time
- Usually completed in 3-15 minutes
- The higher the resolution and the longer the duration, the greater the resource consumption
Use Cases
| Scenario | Role | Example Query | Benefit |
|---|---|---|---|
| Short Video Creation | Creator / Director | "Help me write a cinematic video prompt for a sunset beach" | Professional-grade prompts, generate high-quality video in one go |
| Product Advertising | E-commerce / Brand Operations | "Video prompt for a rotating product display, with a premium feel" | Quickly produce product video materials and reduce filming costs |
| Batch Content Production | Content Operations / MCN | "Give me a set of prompt templates for city night scenes and food" | Reuse templates for batch generation and improve content production efficiency |
| Troubleshooting effect issues | All users | "The people in the video are always distorted, how can I fix this?" | Accurately diagnose the problem and provide actionable fix solutions |
File structure
video-prompt-expert/
├── SKILL.md # 主技能文件（本文档）
├── scripts/
│ └── videogen.py # 视频生成主程序
├── references/
│ └── core_workflow.md # 核心执行流程
└── assets/
└── templates/
└── prompt_templates.json # 结构化模板数据