Higgsfield Supercomputer
kevinbadi/higgsfield-supercomputer
Central de comando para gerações Higgsfield AI com skills pré-carregados
Ver no GitHubDashboard HTML self-contained + bundle de 4 Claude Skills pré-configurados para geração de imagens, vídeos, treinamento de identidades e fotografia de produtos via Higgsfield AI. Clone, autentique com sua chave de API e comece a usar sem dependências externas.
Instalação
npx convez-ai kevinbadi/higgsfield-supercomputer1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Gerar imagens e vídeos através do catálogo completo de modelos Higgsfield
- Treinar identidades de personagens (Soul Character) com face/identity training
- Criar cards de listagem de produtos automaticamente
- Realizar fotossesões de produtos parametrizadas
Veredito

Solução pronta para quem já usa Higgsfield AI e precisa centralizar gerações com interface desktop. Instalação simples (um arquivo HTML + CLI), mas valor limitado fora do ecossistema Higgsfield.
README
A tradução para o português deste README ainda está na fila.
Higgsfield Supercomputer
A personal command center for Higgsfield AI generations — a single-file HTML dashboard paired with a pre-loaded bundle of Higgsfield Claude Skills. Fork it, hook up your Higgsfield API key, and you're running.
What's in this repo
dashboard.html— Self-contained dashboard. No build step, no dependencies. Open it in a browser..agents/skills/— Four Higgsfield Claude Skills, pre-installed:higgsfield-generate— image/video generation across the full Higgsfield model catalog (Soul, Nano Banana, Seedance, Kling, Marketing Studio, etc.)higgsfield-soul-id— Soul Character (face/identity) traininghiggsfield-marketplace-cards— listing card generationhiggsfield-product-photoshoot— product photography
skills-lock.json— version manifest pinning the four skills above tohiggsfield-ai/skillson GitHub.
Quick setup
1. Clone the repo
git clone https://github.com/kevinbadi/higgsfield-supercomputer.git
cd higgsfield-supercomputer
2. Install the Higgsfield CLI
curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh
3. Authenticate
Get a Higgsfield account at higgsfield.ai, then:
higgsfield auth login
Verify with higgsfield account status.
4. Skills
If you're using a runtime that picks up project-level skills from .agents/skills/ (e.g. Claude Code in this directory), the four Higgsfield skills load automatically. Otherwise, install them via your skills manager using skills-lock.json as the manifest.
5. Open the dashboard
open dashboard.html
Customizing the dashboard
dashboard.html ships with all media src="" attributes intentionally blank — your fork starts clean. Drop your own generation URLs (videos, image thumbnails, hero clips) into the src attributes to populate the gallery.
Layout is fully responsive (mobile / tablet / desktop) and the visual system is defined as CSS variables at the top of the <style> block if you want to retheme.
License
No license file ships with this repo by default — add one (MIT is a safe pick) if you plan to distribute beyond personal use.