Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    3d creation

    1

    Professional 3D modeling and WebGL narrative site creation with built-in topology and shader expertise.

    Free

    Security scanned
    3d creation

    3d creation

    Example session with this skill installed

    Build a studio-lit 3D scene in Blender of a brushed metal watch. Use quad-dominant topology and a procedural PBR material for the metal. I have the Blender MCP connected.

    • Read your context and instructions
    • Compiled the 3d creation

    I have initialized the 3D scene in Blender. I focused on quad-dominant topology for the watch casing to ensure clean highlights and applied a custom procedural PBR material for the brushed metal using a noise-driven anisotropy setup.

    3D Scene Specification & Render Log

    AttributeImplementation DetailStatus
    GeometryQuad-only mesh, edge-weighted bevels for the bezelVerified
    MaterialPrincipled BSDF: Metallic 1.0, Roughness 0.25, Procedural Brushed TextureApplied
    LightingThree-point studio setup: Key (Area), Fill (Area), Rim (Spot)Active
    TopologyManifold geometry, no N-gons, circular edge loops for dialChecked
    IterationRound 1: Basic mesh and lightingComplete
    
    # Procedural Brushed Metal Node Setup Snippet
    # Using Noise Texture -> Mapping (Scale 50, 1, 1) -> Bump
    import bpy
    
    def create_brushed_metal():
        mat = bpy.data.materials.new(name="BrushedMetal")
        mat.use_nodes = True
        nodes = mat.node_tree.nodes
        
        bsdf = nodes.get("Principled BSDF")
        bsdf.inputs['Metallic'].default_value = 1.0
        bsdf.inputs['Roughness'].default_value = 0.3
        
        tex = nodes.new('ShaderNodeTexNoise')
        tex.inputs['Scale'].default_value = 500.0
        tex.inputs['Detail'].default_value = 15.0
        
        mapping = nodes.new('ShaderNodeMapping')
        mapping.inputs['Scale'].default_value[1] = 0.1
        # Node connections omitted for brevity
    

    Next steps

    1. Run the render_screenshot tool to verify the anisotropy direction on the metal casing.
    2. Adjust the Rim light intensity if the watch silhouette is not sufficiently separated from the background.
    3. Add the glass crystal using a thin-film interference shader recipe.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    3D production is notoriously difficult for AI agents to get right because code success rarely equals visual success. This skill bridges that gap by providing a deep domain knowledge engine for two specialized workflows: high-end Blender modeling and narrative WebGL websites. It moves beyond simple scripting by embedding principles of topology, anatomy, and PBR materials for Blender, alongside shader patterns and GSAP motion physics for the web.

    What it does

    • Blender modeling builds objects and scenes with proper quad-dominant topology and edge flow for clean deformation.
    • PBR material generation creates procedural texture recipes for realistic metal, glass, and organic surfaces.
    • VFX & physics implements particle systems and simulations within the Blender environment.
    • Narrative web development crafts award-tier websites using Three.js, GSAP, and custom GLSL shaders.
    • Visual iteration enforces a loop of rendering, screenshotting, and visual inspection to ensure output quality.
    • Cinematic composition applies camera movement, lighting, and Gestalt principles to every scene.

    How it works

    1. Identify the workflow by detecting if the task requires Blender MCP for modeling or a web stack for interactive 3D sites.
    2. Apply domain references to pull specific formulas for topology, shaders, or motion easing based on the target style.
    3. Execute and render the initial 3D scene or component code.
    4. Inspect and diff the visual output against the original spec using image-understanding to identify aesthetic or technical flaws.
    5. Iterate through 3-15 rounds of refinements until the visual result meets professional standards.

    Frameworks & tools

    Blender (via MCP), Three.js, GSAP, ScrollTrigger, GLSL/Shaders, Vite, and Lenis.

    Why this beats prompting it yourself

    Standard prompts often produce "dead" 3D scenes with flat lighting and broken topology. This skill embeds specific mathematical recipes for PBR materials and professional cinematography rules that simple instructions cannot replicate.

    Use cases

    • Model a vehicle with Class-A surfaces and studio-quality lighting in Blender.
    • Create an interactive scroll-driven storytelling landing page with WebGL effects.
    • Build a procedural material library for architectural visualizations.
    • Design a high-performance 3D portfolio site that maintains 60fps on mobile.

    Known limitations

    Blender workflows require an active Blender MCP connection. Web tasks are optimized for a Vite/Three.js/GSAP stack and may require adjustment for other frameworks.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.

    Blender modeling requires the Blender MCP to be active. Web outputs are optimized for Three.js, GSAP, and Vite. Iterative visual inspection is required for quality results.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 22 days ago

    • Free with an Agensi account

    Software
    Blender MCP

    Permissions

    Browser
    Read Files
    Write Files
    Network Access
    Listed22 days ago

    What's inside

    Frequently Asked Questions

    Unlike standard prompts that just output code, this skill uses a visual feedback loop to inspect renders, ensuring proper topology, PBR material accuracy, and professional lighting setups.

    The skill is optimized for AI agents capable of using Model Context Protocol (MCP) and can be integrated into any agentic workflow that supports external tool calling and web development stacks.

    Blender workflows require an active Blender MCP connection on your local machine, while WebGL tasks are designed to run within a standard Vite, Three.js, and GSAP development environment.

    You receive the full skill logic, a library of procedural shader recipes, motion easing formulas for GSAP, and a set of domain-specific instructions for handling complex 3D geometry.

    Yes, the skill is specifically designed to handle narrative, scroll-driven websites by combining Three.js scene management with GSAP and ScrollTrigger for high-performance animations.

    The skill specializes in quad-dominant topology, which ensures that models are optimized for clean deformations, high-quality subdivision, and professional-grade rendering.

    New in Agents & Orchestration

    Free