https://github.com/shawnxie94/aether
Codex插件，能将图片沉淀为可复用的视觉记忆，助你持续创作出美学风格一致的图片。The Codex plugin can turn images into reusable visual memories, helping you consistently create images with a unified aesthetic style.
https://github.com/shawnxie94/aether
ai-image-generation codex-plugin codex-skill creative-tools image-generation prompt-refinement visual-memory
Last synced: 16 days ago
JSON representation
Codex插件，能将图片沉淀为可复用的视觉记忆，助你持续创作出美学风格一致的图片。The Codex plugin can turn images into reusable visual memories, helping you consistently create images with a unified aesthetic style.
- Host: GitHub
- URL: https://github.com/shawnxie94/aether
- Owner: shawnxie94
- License: mit
- Created: 2026-05-26T12:58:28.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2026-06-22T05:46:33.000Z (23 days ago)
- Last Synced: 2026-06-22T07:25:25.698Z (23 days ago)
- Topics: ai-image-generation, codex-plugin, codex-skill, creative-tools, image-generation, prompt-refinement, visual-memory
- Language: Python
- Homepage: https://www.npmjs.com/package/aether-codex-plugin
- Size: 6.36 MB
- Stars: 2
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.en.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
README
# Aether
English | [简体中文](README.md)
Aether turns reference images, prompt ideas, and generated results into reusable visual memory, helping you keep creating images with a consistent aesthetic direction.
## Core Capabilities
- **Reusable visual memory:** Extract stable visual language from reference images and generated results, then save it as memory that can be reused over time instead of one-off prompt text.
- **Memory-aware prompt refinement:** Recall saved style, lighting, palette, composition, mood, scene, character, and negative rules, then combine them into a more stable prompt while preserving the user's original intent.
- **Evolvable visual systems:** Decide whether new visual material should become new memory, be added to existing memory, be saved as a variant, or become a merge candidate, so the visual library can grow without becoming chaotic.
- **Generation feedback loop:** Record generated results, visual consistency reviews, and user feedback, so later prompts can reuse what worked and avoid known drift.
- **Natural-language workflow:** Use natural language to complete the full process of capturing, refining, generating, and reusing visual memory.
## Quick Start
Install with npm:
```bash
npx aether-codex-plugin install
```
Verify the local installation:
```bash
aether doctor
```
Restart Codex after installation, or open a new thread so the plugin skills can reload.
## Example Results
> The examples use `gpt-image-2`. Results may vary across image models.
| Reference | Generated |
| --- | --- |
| 
| 
|
| 
| 
|
| 
| 
|
## Usage
After installation, invoke Aether as a Codex plugin with `@Aether`, or as a skill with `$aether-orchestrator`. Aether automatically chooses the right workflow from your natural-language request.
| Workflow | Example |
| --- | --- |
| Capture visual memory | 
|
| Browse visual memory | 
|
| Refine prompts | 
|
| Generate images | 
|
| Edit images | 
|
## License
MIT. See [LICENSE](LICENSE).