gpt image 2 style library
- Repo stars 6,954
- Author repo awesome-gpt-image-2
- Domain
- AI
- Author-declared agents
- No explicit declaration found; this is not inferred or tested compatibility
- Static check
- 88 / 100 · heuristic scan, not runtime safety proof
- Runtime evidence
- Not run; no isolated runtime test evidence is currently recorded
- Author / version / license
- @freestylefly · no license declared
- Token usage
- Lean
- Setup complexity
- Plug-and-play
- External API key
- Not required
- Operating systems
- macOS · Linux · Windows
- Runtime requirements
- No explicit requirements detected
- Detected file/system behavior
-
- Read-only
- Write / modify
- Network behavior
- Local-only
- Install commands
- None (reference only)
Heads up: Tool permissions are not declared; actual access depends on the runtime.
# GPT-Image2 Style Library
Use this skill to turn a user's image-generation intent into a production-ready GPT-Image2 prompt using the awesome-gpt-image-2 style library.
## Example Output

Example request: `用 gpt-image-2-style-library 技能生成城市生命系统图谱`
## Reference
- Read `references/style-library.md` before choosing a template or style.
- The reference is generated from `data/style-library.json` in the repository.
- Prefer the reference over memory when template names, categories, covers, or style tags matter.
## Workflow
1. Detect the user's language and answer in that language.
2. Identify the user's target output: product, poster, UI, infographic, brand, photo, illustration, character, scene, history, document, or special task.
3. Match the request in this order: template category, visual style tag, scene tag, then nearest example cases.
4. If one template is clearly strongest, use it directly. If several are plausible, present 2-3 options with short reasons and ask the user to choose.
5. Build the final prompt with these blocks:
- subject and task
- composition and layout
- visual style and materials
- text and label requirements
- aspect ratio and output format
- constraints and negative details
6. Include the selected template name and any useful example case IDs.
## Output Defaults
- Provide a copyable prompt first.
- Keep constraints concrete: exact text, aspect ratio, readable labels, layout hierarchy, and avoided artifacts.
- For Chinese requests, write the final prompt in Chinese unless the user asks for English.
- For English requests, write the final prompt in English unless the user asks for Chinese.
- When the user asks for multiple concepts, reuse one template and vary subject, composition, palette, and scene.
## Maintenance
When the source repository changes, run:
```bash
npm run generate:style-skill
```
To install the skill into the local Codex skill folder, run:
```bash
npm run install:skill
```
Sections and files
Listed in source order.
sections -> Example Output → Reference → Workflow → Output Defaults → Maintenance
terms -> - Read references/style-library.md before choosing a template or style. · 1. Detect the user's language and answer in that language. · - Provide a copyable prompt first.
files/cmd -> 用 gpt-image-2-style-library 技能生成城市生命系统图谱 · references/style-library.md · data/style-library.json · assets/city-life-system-map.png
body sha256 -> 6c862849e1e3
Reference
Read
references/style-library.mdbefore choosing a template or style. The reference is generated fromdata/style-library.jsonin the repository. Prefer the reference over memory when template names, categories, covers, or style tags matter.Workflow
Detect the user's language and answer in that language. Identify the user's target output: product, poster, UI, infographic, brand, photo, illustration, character, scene, history, document, or special task. Match the request in this order: template category, visual style tag, scene tag, then nearest example cases. If one template is clearly strongest, use it directly. If several are plausible, present 2-3 options with short reasons and ask the user to choose.
Output Defaults
Provide a copyable prompt first. Keep constraints concrete: exact text, aspect ratio, readable labels, layout hierarchy, and avoided artifacts. For Chinese requests, write the final prompt in Chinese unless the user asks for English. For English requests, write the final prompt in English unless the user asks for Chinese. When the user asks for multiple concepts, reuse one template and vary subject, composition, palette, and scene.
Maintenance
When the source repository changes, run: To install the skill into the local Codex skill folder, run: