Back to Skills

create-3d-scene

pjt222
Updated 2 days ago
6 views
17
2
17
View on GitHub
Metaautomationdesigndata

About

This Claude Skill automates Blender 3D scene creation using Python's bpy module. It programmatically configures objects, materials, lighting, cameras, and environments for reproducible visualizations. Use it to automate rendering setups, generate scene variations, or integrate 3D visualization into data pipelines.

Quick Install

Claude Code

Recommended
Primary
npx skills add pjt222/agent-almanac -a claude-code
Plugin CommandAlternative
/plugin add https://github.com/pjt222/agent-almanac
Git CloneAlternative
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/create-3d-scene

Copy and paste this command in Claude Code to install this skill

Documentation

建三維之景

以 Python API(bpy)程式化設 Blender 之全景。配景層級、增網體、建 PBR 材質含節式著色、置光與鏡、設環境與世界。

用時

  • 由無而建可重現之三維視景
  • 自動設產品視或建築渲染
  • 程式化生多景之變
  • 為批渲染建樣景
  • 手修前之原型佈
  • 三維視入數管或報告系

景說配置物、材、光之求產品尺、材色、光設
出求解析、引擎、質1920x1080、Cycles、128 樣
資路外模型、紋、HDRI/path/to/hdri.exrproduct_model.obj
鏡設位、旋、焦、景深location=(7,-7,5)lens=50mm
配置世著色、背、環境HDRI 光、純色、漸層

一、設腳本之構

建 Python 腳本,有正入與構:

#!/usr/bin/env python3
"""
Scene setup script for Blender.
Usage: blender --background --python setup_scene.py
"""

import bpy
import math
import os
from pathlib import Path

def clear_scene():
    """Remove all objects from the scene."""
    bpy.ops.object.select_all(action='SELECT')
    bpy.ops.object.delete(use_global=False)

    # Clear orphaned data
    for block in bpy.data.meshes:
        if block.users == 0:
            bpy.data.meshes.remove(block)

    for block in bpy.data.materials:
        if block.users == 0:
            bpy.data.materials.remove(block)

def main():
    clear_scene()
    # Scene setup steps follow

if __name__ == "__main__":
    main()

得: 腳本有 clear_scene() 與 main() 之構 敗則: 察 Python 語法,驗 bpy 於 Blender Python 境可入

二、增網體

建原始或入網體:

def add_objects():
    """Add mesh objects to scene."""
    # Add cube
    bpy.ops.mesh.primitive_cube_add(
        size=2.0,
        location=(0, 0, 1)
    )
    cube = bpy.context.active_object
    cube.name = "Product_Base"

    # Add sphere
    bpy.ops.mesh.primitive_uv_sphere_add(
        radius=1.0,
        segments=32,
        ring_count=16,
        location=(3, 0, 1)
    )
    sphere = bpy.context.active_object
    sphere.name = "Detail_Sphere"

    # Import external model (optional)
    # bpy.ops.import_scene.obj(filepath="model.obj")

    return cube, sphere

得: 物現於景,名位皆正 敗則: 察算子語法、驗坐標、防名衝

三、以節式著色建材

以著色節設 PBR 材:

def create_material(name, base_color, metallic=0.0, roughness=0.5):
    """Create a PBR material with node setup."""
    # Create material
    mat = bpy.data.materials.new(name=name)
    mat.use_nodes = True
    nodes = mat.node_tree.nodes
    links = mat.node_tree.links

    # Clear default nodes
    nodes.clear()

    # Add Principled BSDF
    node_bsdf = nodes.new(type='ShaderNodeBsdfPrincipled')
    node_bsdf.location = (0, 0)
    node_bsdf.inputs['Base Color'].default_value = base_color + (1.0,)  # Add alpha
    node_bsdf.inputs['Metallic'].default_value = metallic
    node_bsdf.inputs['Roughness'].default_value = roughness

    # Add Material Output
    node_output = nodes.new(type='ShaderNodeOutputMaterial')
    node_output.location = (300, 0)

    # Link nodes
    links.new(node_bsdf.outputs['BSDF'], node_output.inputs['Surface'])

    return mat

def apply_materials(cube, sphere):
    """Apply materials to objects."""
    # Create materials
    mat_red = create_material("RedPlastic", (0.8, 0.1, 0.1), metallic=0.0, roughness=0.4)
    mat_metal = create_material("Metal", (0.8, 0.8, 0.8), metallic=1.0, roughness=0.2)

    # Assign to objects
    if cube.data.materials:
        cube.data.materials[0] = mat_red
    else:
        cube.data.materials.append(mat_red)

    if sphere.data.materials:
        sphere.data.materials[0] = mat_metal
    else:
        sphere.data.materials.append(mat_metal)

得: 材現於著色編,節聯正 敗則: 察節類存,驗聯語法,確色值於 [0,1] 內

四、設光

設光以照景:

def setup_lighting():
    """Add lights to scene."""
    # Sun light
    bpy.ops.object.light_add(
        type='SUN',
        location=(5, 5, 10)
    )
    sun = bpy.context.active_object
    sun.name = "KeyLight"
    sun.data.energy = 3.0
    sun.rotation_euler = (math.radians(45), 0, math.radians(45))

    # Area light (fill light)
    bpy.ops.object.light_add(
        type='AREA',
        location=(-4, -4, 6)
    )
    area = bpy.context.active_object
    area.name = "FillLight"
    area.data.energy = 200.0
    area.data.size = 5.0
    area.rotation_euler = (math.radians(60), 0, math.radians(-135))

    # Point light (rim light)
    bpy.ops.object.light_add(
        type='POINT',
        location=(2, -5, 3)
    )
    point = bpy.context.active_object
    point.name = "RimLight"
    point.data.energy = 500.0

得: 三光,強度與位皆宜 敗則: 依引擎(Cycles 對 EEVEE)調能量,察旋式

五、置鏡

設鏡以正取景:

def setup_camera():
    """Add and configure camera."""
    bpy.ops.object.camera_add(
        location=(7, -7, 5)
    )
    camera = bpy.context.active_object
    camera.name = "MainCamera"

    # Point camera at origin
    direction = (0, 0, 1) - camera.location
    rot_quat = direction.to_track_quat('-Z', 'Y')
    camera.rotation_euler = rot_quat.to_euler()

    # Camera settings
    camera.data.lens = 50  # Focal length in mm
    camera.data.dof.use_dof = True
    camera.data.dof.focus_distance = 10.0
    camera.data.dof.aperture_fstop = 2.8

    # Set as active camera
    bpy.context.scene.camera = camera

得: 鏡置而焦長、景深皆正 敗則: 若 track_to 敗,用簡旋法;驗鏡頭單位

六、設世界之環境

設世之著色與背:

def setup_world():
    """Configure world environment."""
    world = bpy.data.worlds['World']
    world.use_nodes = True
    nodes = world.node_tree.nodes
    links = world.node_tree.links

    # Clear default nodes
    nodes.clear()

    # Add Environment Texture (for HDRI)
    node_env = nodes.new(type='ShaderNodeTexEnvironment')
    node_env.location = (-300, 0)

    # Load HDRI if available
    hdri_path = "/path/to/hdri.exr"
    if os.path.exists(hdri_path):
        node_env.image = bpy.data.images.load(hdri_path)

    # Add Background shader
    node_bg = nodes.new(type='ShaderNodeBackground')
    node_bg.location = (0, 0)
    node_bg.inputs['Strength'].default_value = 1.0

    # Add World Output
    node_output = nodes.new(type='ShaderNodeOutputWorld')
    node_output.location = (300, 0)

    # Link nodes
    links.new(node_env.outputs['Color'], node_bg.inputs['Color'])
    links.new(node_bg.outputs['Background'], node_output.inputs['Surface'])

得: 世著色以 HDRI 或純背設 敗則: 若 HDRI 文件缺,略其加載,獨用 Background 節與色

七、設渲染

設基本渲參:

def setup_render_settings():
    """Configure render settings."""
    scene = bpy.context.scene

    # Render engine
    scene.render.engine = 'CYCLES'  # or 'BLENDER_EEVEE'
    scene.cycles.samples = 128
    scene.cycles.use_denoising = True

    # Output settings
    scene.render.resolution_x = 1920
    scene.render.resolution_y = 1080
    scene.render.resolution_percentage = 100

    # File format
    scene.render.image_settings.file_format = 'PNG'
    scene.render.image_settings.color_mode = 'RGBA'
    scene.render.image_settings.color_depth = '16'
    scene.render.filepath = "/tmp/render_"

得: 渲設已配,可渲 敗則: 察引擎名拼寫,驗解析為正整數

八、組景之層

建集以組:

def organize_collections():
    """Organize objects into collections."""
    # Create collections
    col_geometry = bpy.data.collections.new("Geometry")
    col_lights = bpy.data.collections.new("Lights")
    col_cameras = bpy.data.collections.new("Cameras")

    # Link to scene
    bpy.context.scene.collection.children.link(col_geometry)
    bpy.context.scene.collection.children.link(col_lights)
    bpy.context.scene.collection.children.link(col_cameras)

    # Move objects to collections
    for obj in bpy.data.objects:
        # Unlink from main collection
        bpy.context.scene.collection.objects.unlink(obj)

        # Link to appropriate collection
        if obj.type == 'MESH':
            col_geometry.objects.link(obj)
        elif obj.type == 'LIGHT':
            col_lights.objects.link(obj)
        elif obj.type == 'CAMERA':
            col_cameras.objects.link(obj)

得: 物按名組於諸集,易管 敗則: 建前察集存,處孤物

  • 腳本於 Blender 背景模式無訛而運
  • 諸物現於景概
  • 材於著色編顯正色與性
  • 鏡置而物於框
  • 光足(試渲)
  • 世環正加載(HDRI 或背色)
  • 渲設合出之求
  • 景按集有序
  • 無孤資料(無用之材、網)
  • 腳本含 clear_scene() 以可重現

  1. 物名衝:用唯一之名,建前察物存
  2. 色式誤:RGB 值宜為 (r,g,b,a) 元組,值於 [0,1]
  3. 缺 alpha:設色時含 alpha:(r, g, b, 1.0)
  4. 節聯訛:聯前驗節類有所期之入出
  5. 鏡未激:宜設 bpy.context.scene.camera = camera_object
  6. 相對對絕對路:用絕對路或 Path() 以跨平台
  7. 單位混:Blender 默用米,鏡頭以毫米
  8. 旋式:用 math.radians() 度轉弧度
  9. 引擎異:EEVEE 與 Cycles 功能參數異
  10. 記憶漏:清孤資以防批處積

GitHub Repository

pjt222/agent-almanac
Path: i18n/wenyan/skills/create-3d-scene
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

Related Skills

content-collections

Meta

This skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.

View skill

polymarket

Meta

This skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.

View skill

creating-opencode-plugins

Meta

This skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.

View skill

sglang

Meta

SGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.

View skill