Prompts Index
# Prompts Index
This directory contains **reproducible prompt patterns**
for AI-generated videos used in **education, research, and technical presentations**.
The primary goal is **stability and correctness**,
**not creativity, storytelling, or visual novelty**.
AI-generated video in this repository is treated as **contextual support**,
never as a source of factual information.
---
## Prompt Categories
### 1. Abstract / Mood
Prompts in this category are used to create **non-informational background motion**.
Typical use cases:
- lecture introductions
- presentation openings
- section transitions
- atmosphere setting only
Files:
- `abstract_intro.md`
- `tech_mood.md`
- `research_lab.md`
Constraints:
- no text
- no numbers
- no symbols
- short duration only (typically 3–10 seconds)
- no identifiable objects or readable elements
Notes:
- These videos must not convey meaning by themselves.
- They exist solely to prepare visual context for human-delivered content.
---
### 2. Image to Video
Prompts in this category animate **existing static visuals**
while preserving their informational integrity.
Typical use cases:
- schematic diagrams
- conceptual figures
- static illustrations
- slides converted to subtle motion
Files:
- `image_to_video_basic.md`
Constraints:
- original structure must not change
- geometry must remain intact
- motion must be minimal and non-distracting
- discard output if layout, proportions, or topology are altered
Notes:
- Accuracy originates from the source image.
- The AI model must not invent, simplify, or reinterpret structure.
---
## Generated Samples
The following samples are included as **reference outputs**
to document achievable quality and behavior.
### `abstract_lab_01.mp4`
- Model: Runway **Gen-4 Turbo** (Image → Video)
- Duration: 5 seconds
- Aspect ratio: 16:9
- Visual focus: abstract laboratory environment, blue/white tone
- Motion: slow, stable, cinematic forward movement
- Intended use:
- lecture introduction
- research presentation opening
- neutral technical atmosphere
Location:
samples/output/abstract_lab_01.mp4
---
## General Principles
- AI video provides **context**, not content.
- Information must originate from:
- spoken explanation
- static images
- diagrams
- text or equations outside the video
- If the correctness of an animation cannot be guaranteed, **do not animate**.
- Visual clarity is prioritized over visual appeal.
---
## Status and Maintenance Policy
This prompt set is considered **stable**.
New prompts should be added only if:
- they introduce a clearly new category, or
- they are required by a concrete educational or research use case.
Experimental or exploratory prompts must not be added here.
---
*This directory is intentionally conservative by design.*when to use it
Community prompt sourced from the open-source GitHub repo Samizo-AITL/ai-video-lab (no explicit license). A "Prompts Index" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.
tags
educationcommunitygeneral
source
Samizo-AITL/ai-video-lab · no explicit license