Stability AI — Gerador de Imagens Profissional
Selective Reading Rule
Start with:
references/usage-routing.mdreferences/quality-checklist.md
Then load only the inherited docs, scripts, assets, or examples that match the user's actual task.
Selective Reading Rule
Start with:
references/senior-master-standard.md
Then load only the references, scripts, assets, or examples needed for the user's actual task.
Overview
Geracao de imagens via Stability AI (SD3.5, Ultra, Core). Text-to-image, img2img, inpainting, upscale, remove-bg, search-replace. 15 estilos artisticos.
When to Use This Skill
- When the user mentions "stability ai" or related topics
- When the user mentions "stable diffusion" or related topics
- When the user mentions "sd3.5" or related topics
- When the user mentions "gerar arte" or related topics
- When the user mentions "gerar ilustracao" or related topics
- When the user mentions "image to image" or related topics
Do Not Use This Skill When
- The task is unrelated to stability ai
- A simpler, more specific tool can handle the request
- The user needs general-purpose assistance without domain expertise
How It Works
Skill para gerar imagens artisticas e fotorrealistas usando a Stability AI API.
Gratuito com Community License (sem limite para uso pessoal/pequenas empresas).
Quando Usar Esta Skill Vs Ai-Studio-Image
| Cenario |
Skill recomendada |
| Foto humanizada para Instagram/redes sociais |
ai-studio-image |
| Arte digital, ilustracao, concept art |
stability-ai |
| Foto com camera de celular (realismo casual) |
ai-studio-image |
| Fotorrealismo cinematografico (8K, detalhado) |
stability-ai |
| Material educacional com visual profissional |
ai-studio-image |
| Poster, wallpaper, book cover, game asset |
stability-ai |
| Inpainting (editar parte de uma imagem) |
stability-ai |
| Upscale (aumentar resolucao) |
stability-ai |
| Remover fundo de imagem |
stability-ai |
| Search & Replace (trocar objeto em imagem) |
stability-ai |
| Apagar elemento de uma imagem |
stability-ai |
Setup Rapido
- Criar conta em platform.stability.ai (gratuito)
- Copiar API Key do dashboard
- Colar no
.env: STABILITY_API_KEY=sk-sua-chave-aqui
pip install -r scripts/requirements.txt
Detalhes completos em references/setup-guide.md.
1. Modos De Operacao
| Comando |
O que faz |
Endpoint |
--mode generate |
Texto para imagem (SD3.5) |
/generate/sd3 |
--mode ultra |
Texto para imagem premium |
/generate/ultra |
--mode core |
Texto para imagem rapido |
/generate/core |
--mode img2img |
Imagem + texto para nova imagem |
/generate/sd3 |
--mode upscale |
Aumentar resolucao (conservativo) |
/upscale/conservative |
--mode upscale-creative |
Aumentar resolucao com detalhes |
/upscale/creative |
--mode remove-bg |
Remover fundo (PNG transparente) |
/edit/remove-background |
--mode inpaint |
Editar parte da imagem (mascara) |
/edit/inpaint |
--mode search-replace |
Trocar objeto por descricao |
/edit/search-and-replace |
--mode erase |
Apagar parte da imagem |
/edit/erase |
2. Exemplos De Uso
## Geracao Basica (Sd 3.5 Large)
python scripts/generate.py --prompt "a serene mountain landscape at sunset" --mode generate
## Qualidade Maxima (Ultra)
python scripts/generate.py --prompt "cinematic portrait, dramatic lighting" --mode ultra --aspect-ratio 16:9
## Rapido Para Iteracao (Core)
python scripts/generate.py --prompt "cute cat ninja" --mode core --style anime
## Image-To-Image
python scripts/generate.py --prompt "watercolor style" --mode img2img --image foto.jpg --strength 0.7
## Upscale Conservativo
python scripts/generate.py --prompt "landscape photo" --mode upscale --image foto_pequena.jpg
## Remover Fundo
python scripts/generate.py --mode remove-bg --image produto.jpg
## Inpainting Com Mascara
python scripts/generate.py --prompt "red roses" --mode inpaint --image jardim.jpg --mask mascara.png
## Search & Replace
python scripts/generate.py --prompt "a golden retriever" --mode search-replace --image parque.jpg --search "the cat"
## Apagar Objeto
python scripts/generate.py --mode erase --image foto.jpg --mask area.png
## Listar Modelos
python scripts/generate.py --list-models
## Listar Estilos
python scripts/generate.py --list-styles
## Analisar Prompt (Sugestoes Automaticas)
python scripts/generate.py --prompt "anime warrior girl, widescreen" --analyze --json
3. Aspect Ratios
| Nome |
Ratio |
Aliases |
Uso tipico |
| square |
1:1 |
ig, instagram, quadrado |
Feed Instagram |
| portrait |
2:3 |
retrato, pinterest |
Retrato, poster |
| landscape |
3:2 |
paisagem, horizontal |
Paisagem, banner |
| photo |
4:5 |
ig-feed |
Instagram feed otimizado |
| wide |
16:9 |
widescreen, youtube, cinema, wallpaper |
Cinema, YT |
| ultrawide |
21:9 |
— |
Monitor ultrawide |
| stories |
9:16 |
vertical, tiktok, ig-stories |
Stories, Reels |
| phone |
9:21 |
— |
Wallpaper celular |
4. Estilos (15 Presets)
Cada estilo adiciona qualificadores automaticamente ao prompt:
| Estilo |
Descricao |
Ideal para |
| photorealistic |
Fotorrealismo cinematografico |
Retratos, cenas |
| anime |
Anime/Manga japones |
Personagens, cenas |
| digital-art |
Arte digital detalhada |
Ilustracoes gerais |
| oil-painting |
Pintura a oleo classica |
Arte classica |
| watercolor |
Aquarela fluida |
Arte delicada |
| pixel-art |
Pixel art retro 8/16-bit |
Games retro |
| 3d-render |
Render 3D fotorrealista |
Produtos, cenas 3D |
| concept-art |
Concept art profissional |
Games, filmes |
| comic |
Comics/HQ estilizado |
Quadrinhos |
| minimalist |
Minimalista limpo |
Design, logos |
| fantasy |
Fantasy art epico |
RPG, medieval |
| sci-fi |
Sci-fi futurista |
Cyberpunk, espaco |
| sketch |
Desenho a lapis/carvao |
Estudos, rascunhos |
| pop-art |
Pop art vibrante |
Arte moderna |
| noir |
Film noir dramatico |
Atmosfera sombria |
5. Output
Imagens salvas em data/outputs/ com naming: {mode}_{style}_{timestamp}_{index}.png
Metadados salvos em .meta.json com: prompt original, prompt final, modelo, aspect ratio, seed, tempo, tamanho.
Integracao Com Outras Skills
- ai-studio-image: Complementar — Stability AI para arte, Gemini para fotos humanizadas
- instagram: Gerar arte → publicar no Instagram
- telegram: Gerar imagem → enviar via bot
Rate Limits & Seguranca
- Community License: 150 requests/10 segundos
- Limite diario: 100 imagens/dia (configuravel via
SAFETY_MAX_IMAGES_PER_DAY)
- Retry automatico com backoff exponencial em caso de 429
- Fallback de API keys (primaria + backups)
Referencia De Arquivos
| Arquivo |
Quando consultar |
references/setup-guide.md |
Setup inicial, API key, troubleshooting |
references/prompt-engineering.md |
Tecnicas avancadas de prompt |
references/api-reference.md |
Endpoints, parametros, respostas, erros |
Best Practices
- Provide clear, specific context about your project and requirements
- Review all suggestions before applying them to production code
- Combine with other complementary skills for comprehensive analysis
Common Pitfalls
- Using this skill for tasks outside its domain expertise
- Applying recommendations without understanding your specific context
- Not providing enough project context for accurate analysis
Related Skills
ai-studio-image - Complementary skill for enhanced analysiscomfyui-gateway - Complementary skill for enhanced analysisimage-studio - Complementary skill for enhanced analysis
Limitations
- Use this skill only when the task clearly matches the scope described above.
- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.
1---2name: stability-ai3description: ALWAYS use this when the request matches Stability AI: Geracao de imagens via Stability AI (SD3.5, Ultra, Core).4---56# Stability AI — Gerador de Imagens Profissional78## Selective Reading Rule910Start with:1112- `references/usage-routing.md`13- `references/quality-checklist.md`1415Then load only the inherited docs, scripts, assets, or examples that match the user's actual task.1617## Selective Reading Rule1819Start with:2021- `references/senior-master-standard.md`2223Then load only the references, scripts, assets, or examples needed for the user's actual task.242526## Overview2728Geracao de imagens via Stability AI (SD3.5, Ultra, Core). Text-to-image, img2img, inpainting, upscale, remove-bg, search-replace. 15 estilos artisticos.2930## When to Use This Skill3132- When the user mentions "stability ai" or related topics33- When the user mentions "stable diffusion" or related topics34- When the user mentions "sd3.5" or related topics35- When the user mentions "gerar arte" or related topics36- When the user mentions "gerar ilustracao" or related topics37- When the user mentions "image to image" or related topics3839## Do Not Use This Skill When4041- The task is unrelated to stability ai42- A simpler, more specific tool can handle the request43- The user needs general-purpose assistance without domain expertise4445## How It Works4647Skill para gerar imagens artisticas e fotorrealistas usando a Stability AI API.48**Gratuito** com Community License (sem limite para uso pessoal/pequenas empresas).4950## Quando Usar Esta Skill Vs Ai-Studio-Image5152| Cenario | Skill recomendada |53|---------|-------------------|54| Foto humanizada para Instagram/redes sociais | ai-studio-image |55| Arte digital, ilustracao, concept art | **stability-ai** |56| Foto com camera de celular (realismo casual) | ai-studio-image |57| Fotorrealismo cinematografico (8K, detalhado) | **stability-ai** |58| Material educacional com visual profissional | ai-studio-image |59| Poster, wallpaper, book cover, game asset | **stability-ai** |60| Inpainting (editar parte de uma imagem) | **stability-ai** |61| Upscale (aumentar resolucao) | **stability-ai** |62| Remover fundo de imagem | **stability-ai** |63| Search & Replace (trocar objeto em imagem) | **stability-ai** |64| Apagar elemento de uma imagem | **stability-ai** |6566## Setup Rapido67681. Criar conta em **platform.stability.ai** (gratuito)692. Copiar API Key do dashboard703. Colar no `.env`: `STABILITY_API_KEY=sk-sua-chave-aqui`714. `pip install -r scripts/requirements.txt`7273Detalhes completos em `references/setup-guide.md`.7475## 1. Modos De Operacao7677| Comando | O que faz | Endpoint |78|---------|-----------|----------|79| `--mode generate` | Texto para imagem (SD3.5) | `/generate/sd3` |80| `--mode ultra` | Texto para imagem premium | `/generate/ultra` |81| `--mode core` | Texto para imagem rapido | `/generate/core` |82| `--mode img2img` | Imagem + texto para nova imagem | `/generate/sd3` |83| `--mode upscale` | Aumentar resolucao (conservativo) | `/upscale/conservative` |84| `--mode upscale-creative` | Aumentar resolucao com detalhes | `/upscale/creative` |85| `--mode remove-bg` | Remover fundo (PNG transparente) | `/edit/remove-background` |86| `--mode inpaint` | Editar parte da imagem (mascara) | `/edit/inpaint` |87| `--mode search-replace` | Trocar objeto por descricao | `/edit/search-and-replace` |88| `--mode erase` | Apagar parte da imagem | `/edit/erase` |8990## 2. Exemplos De Uso9192```bash9394## Geracao Basica (Sd 3.5 Large)9596python scripts/generate.py --prompt "a serene mountain landscape at sunset" --mode generate9798## Qualidade Maxima (Ultra)99100python scripts/generate.py --prompt "cinematic portrait, dramatic lighting" --mode ultra --aspect-ratio 16:9101102## Rapido Para Iteracao (Core)103104python scripts/generate.py --prompt "cute cat ninja" --mode core --style anime105106## Image-To-Image107108python scripts/generate.py --prompt "watercolor style" --mode img2img --image foto.jpg --strength 0.7109110## Upscale Conservativo111112python scripts/generate.py --prompt "landscape photo" --mode upscale --image foto_pequena.jpg113114## Remover Fundo115116python scripts/generate.py --mode remove-bg --image produto.jpg117118## Inpainting Com Mascara119120python scripts/generate.py --prompt "red roses" --mode inpaint --image jardim.jpg --mask mascara.png121122## Search & Replace123124python scripts/generate.py --prompt "a golden retriever" --mode search-replace --image parque.jpg --search "the cat"125126## Apagar Objeto127128python scripts/generate.py --mode erase --image foto.jpg --mask area.png129130## Listar Modelos131132python scripts/generate.py --list-models133134## Listar Estilos135136python scripts/generate.py --list-styles137138## Analisar Prompt (Sugestoes Automaticas)139140python scripts/generate.py --prompt "anime warrior girl, widescreen" --analyze --json141```142143## 3. Aspect Ratios144145| Nome | Ratio | Aliases | Uso tipico |146|------|-------|---------|-----------|147| square | 1:1 | ig, instagram, quadrado | Feed Instagram |148| portrait | 2:3 | retrato, pinterest | Retrato, poster |149| landscape | 3:2 | paisagem, horizontal | Paisagem, banner |150| photo | 4:5 | ig-feed | Instagram feed otimizado |151| wide | 16:9 | widescreen, youtube, cinema, wallpaper | Cinema, YT |152| ultrawide | 21:9 | — | Monitor ultrawide |153| stories | 9:16 | vertical, tiktok, ig-stories | Stories, Reels |154| phone | 9:21 | — | Wallpaper celular |155156## 4. Estilos (15 Presets)157158Cada estilo adiciona qualificadores automaticamente ao prompt:159160| Estilo | Descricao | Ideal para |161|--------|-----------|-----------|162| photorealistic | Fotorrealismo cinematografico | Retratos, cenas |163| anime | Anime/Manga japones | Personagens, cenas |164| digital-art | Arte digital detalhada | Ilustracoes gerais |165| oil-painting | Pintura a oleo classica | Arte classica |166| watercolor | Aquarela fluida | Arte delicada |167| pixel-art | Pixel art retro 8/16-bit | Games retro |168| 3d-render | Render 3D fotorrealista | Produtos, cenas 3D |169| concept-art | Concept art profissional | Games, filmes |170| comic | Comics/HQ estilizado | Quadrinhos |171| minimalist | Minimalista limpo | Design, logos |172| fantasy | Fantasy art epico | RPG, medieval |173| sci-fi | Sci-fi futurista | Cyberpunk, espaco |174| sketch | Desenho a lapis/carvao | Estudos, rascunhos |175| pop-art | Pop art vibrante | Arte moderna |176| noir | Film noir dramatico | Atmosfera sombria |177178## 5. Output179180Imagens salvas em `data/outputs/` com naming: `{mode}_{style}_{timestamp}_{index}.png`181182Metadados salvos em `.meta.json` com: prompt original, prompt final, modelo, aspect ratio, seed, tempo, tamanho.183184## Integracao Com Outras Skills185186- **ai-studio-image**: Complementar — Stability AI para arte, Gemini para fotos humanizadas187- **instagram**: Gerar arte → publicar no Instagram188- **telegram**: Gerar imagem → enviar via bot189190## Rate Limits & Seguranca191192- **Community License**: 150 requests/10 segundos193- **Limite diario**: 100 imagens/dia (configuravel via `SAFETY_MAX_IMAGES_PER_DAY`)194- **Retry automatico** com backoff exponencial em caso de 429195- **Fallback de API keys** (primaria + backups)196197## Referencia De Arquivos198199| Arquivo | Quando consultar |200|---------|-----------------|201| `references/setup-guide.md` | Setup inicial, API key, troubleshooting |202| `references/prompt-engineering.md` | Tecnicas avancadas de prompt |203| `references/api-reference.md` | Endpoints, parametros, respostas, erros |204205## Best Practices206207- Provide clear, specific context about your project and requirements208- Review all suggestions before applying them to production code209- Combine with other complementary skills for comprehensive analysis210211## Common Pitfalls212213- Using this skill for tasks outside its domain expertise214- Applying recommendations without understanding your specific context215- Not providing enough project context for accurate analysis216217## Related Skills218219- `ai-studio-image` - Complementary skill for enhanced analysis220- `comfyui-gateway` - Complementary skill for enhanced analysis221- `image-studio` - Complementary skill for enhanced analysis222223## Limitations224- Use this skill only when the task clearly matches the scope described above.225- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.226- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.
anhvu1107