Skip to content
Content
Skill

/02-3d-cgi

Generate 3D CGI and rendered video prompts for Seedance 2.0 on Higgsfield. Use whenever the user wants 3D rendered, CGI, Pixar-style, Unreal Engine, photorealistic 3D, computer-generated, or digitally rendered video content. Triggers on: 3D animation, CGI, rendered, Blender,

From plugin
higgsfield-seedance2-jineng
74130 skills
Install
$ npx -y skills add beshuaxian/higgsfield-seedance2-jineng --skill 02-3d-cgi --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/02-3d-cgi

Context preview

The summary Claude sees to decide when to auto-load this skill.

Generate 3D CGI and rendered video prompts for Seedance 2.0 on Higgsfield. Use whenever the user wants 3D rendered, CGI, Pixar-style, Unreal Engine, photorealistic 3D, computer-generated, or digitally rendered video content. Triggers on: 3D animation, CGI, rendered, Blender,

SKILL.md

02-3d-cgi.SKILL.md
name: seedance-3d-cgi
description: Generate 3D CGI and rendered video prompts for Seedance 2.0 on Higgsfield. Use whenever the user wants 3D rendered, CGI, Pixar-style, Unreal Engine, photorealistic 3D, computer-generated, or digitally rendered video content. Triggers on: 3D animation, CGI, rendered, Blender, Unreal Engine, octane render, ray tracing, volumetric, subsurface scattering, physically based rendering, or any 3D/CG video request. Always use even if the user just says "make it look 3D" or describes a rendered aesthetic.

Seedance 2.0 on Higgsfield — 3D CGI Style Prompt Generator

Overview

This skill guides you through creating world-class 3D CGI and rendered video prompts optimized for **Seedance 2.0 on Higgsfield**. Whether you're targeting Pixar-style animation, photorealistic product visualization, architectural walkthroughs, sci-fi environments, or creature animation, this framework provides the terminology, structure, and psychology needed to generate compelling CGI content.

The difference between a generic "make it look 3D" request and a production-ready CGI prompt lies in specificity: render engine terminology, material properties, lighting language, camera dynamics, and the architectural "hooks" that grab attention in the first two seconds. This skill teaches all of it.

---

What Seedance 2.0 on Higgsfield Accepts

**Seedance 2.0 on Higgsfield** is a generative video platform optimized for high-quality rendered content. It excels at:

  • **Render Engines & Styles**: Blender Cycles, Unreal Engine 5 (UE5), Octane Render, Arnold, RenderMan, V-Ray, Redshift, Corona Renderer, real-time ray tracing, NVIDIA OptiX
  • **Resolution & Frame Rate**: Up to 4K at 60fps or higher; supports cinematic 24fps and 30fps workflows
  • **Animation & Motion**: Smooth procedural motion, physics simulation, creature rigging, character animation, particle systems, fluid dynamics
  • **Lighting Precision**: HDRI illumination, physically accurate light simulation, global illumination (GI), caustics, volumetric lighting, subsurface scattering (SSS), caustic refraction
  • **Material Complexity**: Procedural texturing, normal/displacement mapping, roughness/metallic workflows, anisotropic surfaces, fabric simulation, translucency, iridescence
  • **Volumetrics & Particles**: Smoke, fire, fog, dust, rain, snow, magical effects, data visualization particles
  • **Camera Sophistication**: Depth of field (DOF), motion blur, chromatic aberration, lens distortion, tracking shots, crane movements, smooth interpolation between cameras
  • **Duration**: Typically 2–10 seconds for optimal impact; can handle longer narrative sequences

**Key Constraint**: Prompts must be visual-first and technically descriptive. Vague requests ("make it look cool") are rejected. Specific render terminology significantly improves output quality.

---

The 2-Second Hook Framework for 3D CGI

Every successful CGI video begins with an immediate visual hook—something that stops the viewer's eye in the first two seconds. The subsequent 8 seconds develop the hook into a complete narrative or reveal.

10+ Hook Techniques for 3D CGI

1. **Dramatic Camera Fly-Through**

The camera moves through impossible geometry at high speed. Used for: sci-fi spaces, architectural visualization, abstract environments, tunnel sequences. Example: "Camera hurtles through a crystalline tunnel, refracting light into fractals as we move."

2. **Scale Reveal (Tiny to Massive)**

Start with an extreme closeup—a water droplet, a grain of sand, a single fiber—then pull back to reveal it's part of something enormous. Example: "Begin on a single raindrop on a leaf. Pull back to reveal a forest canopy in a thunderstorm. Pull back further: a continent being drenched."

3. **Impossible Camera Angle**

Utilize camera positions that human cameras can't occupy: inside glass, on the surface tension of water, rotating through a solid object's cross-section, or orbiting an object impossibly fast. Example: "Camera rotates around a spinning hourglass, rotating through its surface to show the sand inside in realtime."

4. **Object Materializing from Particles**

Begin with thousands of small particles (dust, light, pixels, data) that coalesce into a recognizable 3D object. Example: "Golden particles swirl and coalesce into the shape of a dancing figure, settling into a detailed humanoid form."

5. **Photorealistic Object in Surreal Setting**

Place a hyper-real object (like a perfect glass sphere or a photographic product shot) in an impossible or dreamlike environment. Example: "A pristine crystal crown sits alone on a floor of clouds while a aurora borealis swirls overhead."

6. **Transformative Morph Sequence**

One object smoothly transitions into another, often defying physics. Example: "A marble rolls across white space, morphing mid-motion into a silver sphere, then into a water droplet."

7. **Revealing Through Transparency**

Start with a solid opaque object, then it becomes transparent or reflective, revealing internal structure. Example: "A stone statue begins to crack, light pouring through the fissures. The exterior dissolves to reveal a glowing core of molten light within."

8. **Depth Shift with Depth of Field**

Begin with extreme DOF (background/foreground blurred). Shift focus to reveal background, midground, or foreground in crisp detail while other planes blur. Example: "Focus on a dust particle suspended in light. Shift focus to reveal a vast interior architecture behind it."

9. **Chromatic Separation & Refraction**

Light bends, splits, or refracts, creating rainbow effects and optical distortion. Often used for luxury goods, technology, or abstract sequences. Example: "A prism catches sunlight, fragmenting the entire frame into a spectrum. The light gradually refracts around a product."

10. **Kinetic Decomposition**

A structured object (building, sculpture, machine) disassembles into floating comp

Read more
Ships withhiggsfield-seedance2-jineng

15 个专业 Claude 技能,专为 Seedance 2.0(Higgsfield)打造。每个技能将 Claude 变成特定视频风格的提示词工程师——生成大型、详细、可直接使用的提示词,包含强力 2 秒钩子,让观众停止滑动。 15 specialized Claude skills built for Seedance 2.0 on Higgsfield.

Get the whole plugin
Stats
741
Stars
150
Forks
Maintained
Maintenance
Shell
Language
4mo ago
Last commit
4mo ago
Created

Repo: beshuaxian/higgsfield-seedance2-jineng