Core Idea
The author's design philosophy, what sets this apart from peer projects, and quick notes on the key workflows. Skill names mentioned in the text auto-link to their detail pages.
Higgsfield AI Skills is a collection of agent skills that wrap the Higgsfield AI generation platform into Markdown-based skills for coding agents (Claude Code, Cursor, Codex). It follows a modular, chainable design: each skill does one generation or build task and can be composed with others.
Train Soul → use ref in generate; photoshoot/cards standalone; websites embed generated assets. Agents load skills by user slash command; no auto-load described.
Skill Composition
Each skill is designed for a specific kind of caller — some only fire when you name them, others load automatically when the model judges them relevant. The split below summarises both modes at a glance.
These skills only fire when you call them by name. The agent will not load them on its own — they stay inert until invoked.
These skills are loaded automatically by the agent when it judges the context relevant. You usually don't need to call them yourself.
Skill List
Creates coherent playable web games or game assets via Higgsfield CLI. Method: route request, plan with design system, generate assets, build ZIP, verify locally, deploy for URL. Value: delivers verified playable game URL or usable assets with optional marketplace publish.
Wraps the higgsfield CLI to submit image/video/3D/audio generation and video virality analysis jobs. Provides model selection guides, media flag usage, and Marketing Studio workflows. Outputs asset URLs or analysis reports, enabling fast branded content and media production.
Creates marketplace-ready product visuals via the `higgsfield marketplace-cards create` CLI. It selects scope bundles or custom assets, sends requests to a private backend enhancer, and runs image jobs. Output is ready-to-use image URLs with labels for e-commerce listings.
Generates brand/product images via the `higgsfield product-photoshoot create` CLI. It selects from 10 modes, runs a short interview, and submits to a backend enhancer + gpt_image_2. Output is a list of image URLs for e-commerce, ads, and social media.
Trains a face-faithful Higgsfield Soul identity model from 5–20 photos via the higgsfield CLI. Provides bootstrap, photo collection, variant selection, silent training wait, and reusable id delivery. Enables consistent face across all Soul-powered image and video generations.
Guides Agent to build non-photorealistic explainer videos via Higgsfield CLI by locking one style, writing per-block narration and visual prompts, generating voices then clips, and assembling via explainer_video. Delivers a consistent styled MP4.
Builds standalone brand websites or Higgsfield-integrated AI apps via the Higgsfield CLI and local code editing. Drives create/design/deploy flows with type-specific rules and required cover metadata. Outputs a deployed SSR Cloudflare Worker site and optional community publish.
Installation
Copy and run in your terminal to add higgsfield-ai/skills to your Agent environment.
npx skills add higgsfield-ai/skills