blender-hair-grooming
Skill1.2k starsChanged 3 days agoMITView on GitHub
Hair and fur in Blender 5.3 - hair curves on a surface, guides + interpolated children with the bundled hair node assets (clump, curl, noise, droop), density masks, Principled Hair BSDF (melanin), Curves Sculpt combing, 5.2 node Hair Dynamics, game hair cards with a rendered strand atlas, Alembic/USD groom export. Use for hair, fur, beards, eyebrows, grass tufts.
Trending
More talk lately.about the whole repository
Who's talking about EL4CTEO/blender-skills
- sudosubin/agents.nix“claude-code-marketplaces.github.el4cteo.blender-skills: init at 1.3.1”GitHub · Nov 16, 2026
- sudosubin/agents.nix“agent-skills.github.el4cteo.blender-skills: 1.3.0 → 1.3.1”GitHub · Nov 13, 2026
- sudosubin/agents.nix“agent-skills.github.el4cteo.blender-skills: 1.2.0 → 1.3.0”GitHub · Nov 13, 2026
- akhmadilham/pinax-registry“[Skill Proposal]: blender-skills”GitHub · Nov 3, 2026
- sudosubin/agents.nix“agent-skills.github.el4cteo.blender-skills: 1.1.0 → 1.2.0”GitHub · Oct 30, 2026
The biggest 5 of 18 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 18
Talked about on GitHub (18 times).
On GitHub · 18
- Nov 16, 2026claude-code-marketplaces.github.el4cteo.blender-skills: init at 1.3.1 — sudosubin/agents.nix0 reactions
- Nov 13, 2026agent-skills.github.el4cteo.blender-skills: 1.3.0 → 1.3.1 — sudosubin/agents.nix0 reactions
- Nov 13, 2026agent-skills.github.el4cteo.blender-skills: 1.2.0 → 1.3.0 — sudosubin/agents.nix0 reactions
- Nov 3, 2026[Skill Proposal]: blender-skills — akhmadilham/pinax-registry0 reactions · 1 comment
- Oct 30, 2026agent-skills.github.el4cteo.blender-skills: 1.1.0 → 1.2.0 — sudosubin/agents.nix0 reactions
13 more on GitHub
- Oct 19, 2026feat(agents): install blender-skills globally for all agents — zigrivers/mac-launchpad0 reactions
- Oct 18, 2026[Candidate] el4cteo/blender-skills/skills/blender-hardsurface-modeling - blender-hardsurface-modeling — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 18, 2026Mac Codex: audit and learn upstream Blender Agent Skills with isolated pilot — hengqi61-maker/muscle-atlas0 reactions · 2 comments
- Oct 17, 2026[Candidate] el4cteo/blender-skills/skills/blender-geometry-nodes - blender-geometry-nodes — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 17, 2026[Candidate] el4cteo/blender-skills/skills/blender-game-optimization - blender-game-optimization — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 17, 2026[Candidate] el4cteo/blender-skills/skills/blender-facial-shapekeys - blender-facial-shapekeys — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 16, 2026[Candidate] el4cteo/blender-skills/skills/blender-export-pipelines - blender-export-pipelines — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 16, 2026[Candidate] el4cteo/blender-skills/skills/blender-core - blender-core — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 16, 2026[Candidate] el4cteo/blender-skills/skills/blender-character-rigging - blender-character-rigging — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 16, 2026agent-skills.github.el4cteo.blender-skills: init at 1.1.0 — sudosubin/agents.nix0 reactions
- Oct 16, 2026[Candidate] el4cteo/blender-skills/skills/blender-asset-qa - blender-asset-qa — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 16, 2026[Candidate] el4cteo/blender-skills/skills/blender-animation - blender-animation — ProSkillsMD/proskills0 reactions · 1 comment
- Oct 16, 2026[Candidate] el4cteo/blender-skills/skills/blender-addon-development - blender-addon-development — ProSkillsMD/proskills0 reactions · 1 comment
In this repository29 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- blender-addon-developmentSkill
- blender-animationSkill
- blender-asset-qaSkill
- blender-character-riggingSkill
- blender-clothingSkill
- blender-coreSkill
- blender-environment-kitsSkill
- blender-export-pipelinesSkill
- blender-facial-shapekeysSkill
- blender-game-optimizationSkill
- blender-geometry-nodesSkill
- blender-hardsurface-modelingSkill
- blender-lighting-lookdevSkill
- blender-pbr-materialsSkill
- blender-pipeline-automationSkill
- blender-renderingSkill
- blender-retopology-uvSkill
- blender-scene-setupSkill
- blender-sculptingSkill
- blender-simulationSkill
- blender-stylized-nprSkill
- blender-texture-bakingSkill
- blender-texturingSkill
- blender-time-of-daySkill
- blender-video-editingSkill
- blender-weather-fxSkill
- blender-workflow-recipesSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 27
Look-dev under the final lights.
- Very specific: line 54
No UVs on the surface means Generate has nothing to attach to.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Hair grooming
The execution model is in blender-core. Hair physics details are in blender-simulation, and look-dev in blender-lighting-lookdev.
One call each
scripts/groom.py -- --surface Head --mask Scalp --length 0.15 --children 150000 --clump 0.5
scripts/groom.py -- --surface Head --mask Beard --length 0.02 --droop 0.2 --curl 0.004 --melanin 0.9 --name Beard
scripts/groom.py -- --surface Body --length 0.025 --droop 0.3 --clump 0.3 --name Fur
scripts/groom.py -- --cards Hair --card-count 150 --atlas-res 2048 --out //textures # game hair cardsHow pros groom
- Surface first: the scalp is a mesh (or the head itself) with UVs (hair attaches by UV) and a vertex group painted where hair grows. That group becomes the density mask, so soft edges give a natural hairline.
- Guides → children: a few hundred to a few thousand guides carry the shape, and thousands of interpolated children give the volume. Style the guides, never the children.
- Shape in layers: length, then clump (strands gather into locks), then curl/wave, then frizz/noise, then profile (thin tips). Keep each a separate modifier you can toggle.
- Comb by hand for anything hero: Curves Sculpt mode brushes (
Comb,Add,Puff,Pinch,Grow/Shrink,Smooth,Slide,Density,Delete,Pull Tips). Procedural droop gives a believable base, not a hairstyle. - Look-dev under the final lights. Hair reads mostly from specular and depth (clumps catch highlights), not albedo.
groom.py
- Runs 5.2's Add Empty Hair (
object.curves_empty_hair_add). It parents to the surface, setssurface/surface_uv_map, adds Capture Rest Geometry to the surface, and adds the node Hair Dynamics modifier (pinned last, experimental). Then it stacks the bundled assets: Generate (--densityguides/m²) → Interpolate (--children/m²,--viewportfraction in the viewport) → Clump → Curl → Displace + Shrinkwrap (--droop× length down, kept 2 mm above the surface) → Noise → Set Hair Curve Profile (--radius, tip taper). - Clump/curl guide distance = 10% of
--length. The asset's default 0.1 m collapses a whole head into a few tufts. - Each asset stays a modifier with live inputs, so tweak there. Inputs by name:
getattr(md.properties.inputs, identifier).value; vertex-group masks use.type = "ATTRIBUTE",.attribute_name. - It reports curves, mean length, the max root gap to the surface (must be ~0) and the modifier stacks. Evaluated counts in background mode use the viewport amount.
Numbers
| Value | |
| Human scalp | ~100–150k hairs, ~200/cm². CG: 30k–150k rendered strands, 300–3000 guides |
| Strand radius | real 0.03–0.05 mm. Use 0.05 mm (--radius 0.00005), thicker with fewer strands |
| Melanin (Principled Hair, Chiang) | black 1.0 · dark brown 0.8 · brown 0.65 · light brown 0.45 · blond 0.25 · platinum 0.1. Red: 0.5 + redness 0.9. Grey: 0 + tint |
| Roughness | 0.25–0.35 (glossy) … 0.5 (dry). Radial roughness 0.3 |
| Fur | length 1–5 cm, density 200k–2M/m², little clump |
HUANG (4.0+) models elliptical cross-sections and is better for close-ups. CHIANG is the reference default. |
Game hair cards (--cards HAIR)
- Picks
--card-countwell-spread strands (farthest-point sampling) of the evaluated groom and builds a ribbon along each:--card-widthm, tapering to 30% at the tip, lying flat on the surface (side = tangent × surface normal). UV v runs root 0 → tip 1, one atlas column per card (--atlas-columns). - Renders the atlas with Cycles from flat strand clumps:
T_<name>_BC(darker roots, coverage in alpha, Standard view), plusT_<name>_RootTipandT_<name>_ID(Raw, linear data). The card material uses Dithered alpha with no backface culling. - Pros layer cards from the scalp outward: dense, wide, opaque cards on the scalp to hide skin; mid layers; then thin flyaway cards on top. Budgets: hero 20k–60k tris, NPC 5k–15k. Bend cards in S/C shapes, never flat planes, and check from every angle. Engines use alpha test/dither + depth prepass, root-tip for darkening/AO and ID for per-strand variation. Add a tangent (flow) map along V for anisotropic specular.
Export grooms
Strand hair for UE Groom/Houdini: Alembic (wm.alembic_export) or USD (wm.usd_export(export_curves=True)). Both write the evaluated curves (measured: 246/246 curves round-trip). Bake the modifier stack first if the target needs guides only.
Gotchas
- No UVs on the surface means
Generatehas nothing to attach to. Add UVs before grooming. - Changing the surface topology after grooming breaks attachment. Groom last, or re-run.
- Scale: apply the head's scale. Lengths and densities are world meters.
- Viewport: keep
--viewport≤ 0.2 on full heads. Final strand counts only appear in renders. - Hair Dynamics simulates on playback. For stills on frame 1 it's a pass-through. Bake it (
blender-simulation) before animated renders.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

