Blender Stylized Npr logo

Blender Stylized Npr

Community
EL4CTEO
blender-stylized-npr

Stylized / NPR looks in Blender 5.2 - toon and cel shading (bands, shadow tint, rim, hard specular), anime faces with custom normals, inverted-hull outlines with smoothed outline normals for game engines, Line Art and Freestyle lines, Grease Pencil drawing, unlit export. Use for anime, cartoon, comic, cel-shaded or outlined styles.

Overview

PublisherEL4CTEO
Repositoryblender-skills
Skill nameblender-stylized-npr
Stars
605
Forks
52
Bundled files
2
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by EL4CTEO on GitHub. Read the source before you install it.

Installation

Install the Blender Stylized Npr AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/EL4CTEO/blender-skills/tree/main/skills/blender-stylized-npr
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/EL4CTEO/blender-skills.git /tmp/blender-skills
mkdir -p .claude/skills
cp -r /tmp/blender-skills/skills/blender-stylized-npr .claude/skills/blender-stylized-npr
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Blender Stylized Npr in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Blender Stylized Npr on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Blender Stylized Npr is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Stylized / NPR

The execution model is in blender-core. Base colors still come from blender-pbr-materials/blender-texturing. This skill replaces how they're lit.

One call each

scripts/toon_setup.py -- --objects Hero Sword --bands 2 --rim 0.4 --outline 0.004
scripts/toon_setup.py -- --objects Hero --bands 3 --spec 0.3 --smooth-normals      # + engine outline data
scripts/toon_setup.py -- --objects Hero --lighting scene                           # EEVEE: all lights + cast shadows
scripts/toon_setup.py -- --line-art --outline 0                                    # scene-wide Line Art
scripts/toon_setup.py -- --objects Hero Sword --remove --line-art                  # exact restore

toon_setup.py

  • Inserts an NPR_Toon emission group between each material's Principled base color and the output, and records the original surface link for --remove. Re-running updates values in place.
  • --lighting key (default, Cycles + EEVEE): shading = N·L against one sun (--light; NPR_Key is created if there's none). Drivers follow the sun's rotation, so rotate the sun to art-direct the shadow shape. This is how engine toon shaders work, and the look matches after export.
  • --lighting scene: Shader to RGB on a white diffuse, so every light and cast shadow shapes the bands. EEVEE only. Measured: in Cycles, Shader to RGB returns a constant regardless of light (identical render at 0, 3 and 10 W/m² sun).
  • Bands: 2 (lit/shadow, classic anime) or 3 (adds a mid band 0.45 above --threshold). --softness (0.02) is the half-width of the band edge, so 0 aliases. --shadow-tint multiplies the base color: use cool and saturated (0.6, 0.55, 0.75), never grey/black, because grey shadows read as dirty.
  • --rim (0–1) adds a lit-side rim, and --spec (0–1) a hard Blinn highlight disc (0.2–0.4 for hair/metal, 0 for skin).
  • Sets view transform Standard: AgX/Filmic desaturate and compress flat colors, so the albedo you pick is the pixel you get only under Standard. The original is restored on --remove.

Outlines (pros use several at once)

MethodWhere it worksNotes
Inverted hull (--outline, m)Cycles, EEVEE, enginesSolidify outward, flipped normals, outline slot, Backfacing→transparent. Width 2–6 mm on a 1.75 m character. --outline-group = vertex group for width (taper toward tips, 0 inside the mouth)
--smooth-normalsengine outline shadersNPR_OutlineNormal corner color = position-welded normal in tangent space (0.5 + 0.5n). Hard edges/UV splits no longer open gaps. Decode n = TBN·(c·2−1)
Line Art (--line-art, Grease Pencil)renderscontour + crease + intersection from the camera. Needs a scene camera. --line-radius (m)
Freestylerendersscene.render.use_freestyle = True, line_thickness px. Slowest, most options
Known hull limits: lines at intersections and inner creases are missing (use Line Art for those in renders), and hard-edged meshes split the hull. In Blender, merge by distance before the Solidify. In engines, use --smooth-normals.

Anime faces and hair

  • Face shadows from real normals look blotchy. Transfer custom normals from a simple ellipsoid proxy: a Data Transfer modifier with use_loop_data=True, data_types_loops={'CUSTOM_NORMAL'}, loop_mapping='POLYINTERP_NEAREST' (no auto smooth needed since 4.1), masked by a face vertex group. Engines import the custom normals (FBX/glTF export normals).
  • For a stylized nose/cheek shadow, many games use an SDF face shadow texture sampled with the light angle in-engine. Author it as a grayscale threshold map, not in the node tree.
  • Hair: a strong --spec plus a hand-painted highlight band ("angel ring") in the base color texture.

Grease Pencil (5.x API)

python
gp = bpy.data.grease_pencils.new("Ink"); ob = bpy.data.objects.new("Ink", gp); scene.collection.objects.link(ob)
mat = bpy.data.materials.new("Ink"); bpy.data.materials.create_gpencil_data(mat); gp.materials.append(mat)
drawing = gp.layers.new("Lines").frames.new(1).drawing
drawing.add_strokes([3]); s = drawing.strokes[0]
for i, p in enumerate(s.points): p.position, p.radius = (i * 0.5, 0, 0), 0.01   # radius in m

Strokes expose cyclic, material_index, fill_id, hide_stroke. Raw attributes: drawing.attributes (position, radius). GreasePencilv3/grease_pencils_v3 are gone (5.0). 5.2 Line Art can emit fill strokes (fill_strokes).

To game engines

Node toon shading doesn't export. Export the base color textures and rebuild the toon ramp in the engine (UE material/post-process, Unity URP toon shader, Godot ShaderMaterial), driven by the same sun direction. Ship the hull as a second material slot (apply the modifier), or as NPR_OutlineNormal for a shader outline pass. For flat, unlit assets in glTF, use a Background shader, or Light Path Is Camera Ray mixing Transparent→Background: both export as KHR_materials_unlit. Emission exports as emissive on a black base, not unlit (measured).

Gotchas

  • Apply scale before --outline: Solidify thickness is in object space.
  • Bloom and AO fight the flat look. Keep the world dim and uniform, and use one key light plus maybe a rim light.
  • --lighting scene bands shift with light energy (luminance 0.3 ≈ threshold 0). Raise/lower --threshold instead of the light power.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Blender Stylized Npr AI skill do?

Stylized / NPR looks in Blender 5.2 - toon and cel shading (bands, shadow tint, rim, hard specular), anime faces with custom normals, inverted-hull outlines with smoothed outline normals for game engines, Line Art and Freestyle lines, Grease Pencil drawing, unlit export. Use for anime, cartoon, comic, cel-shaded or outlined styles.

Why use Blender Stylized Npr on TypingMind?

Because you install it once and use it with any model. Blender Stylized Npr is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Blender Stylized Npr in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/EL4CTEO/blender-skills/tree/main/skills/blender-stylized-npr. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Blender Stylized Npr?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Blender Stylized Npr?

As many as you like. As long as a model supports skills, you can use Blender Stylized Npr with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Blender Stylized Npr AI skill free?

Yes. It is published on GitHub by EL4CTEO under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇