EL4CTEOEL4CTEO/blender-skills/AGENTS.md

blender-skills

AGENTS.md779 starsChanged 14 days agoMITView on GitHub

This repo contains Agent Skills (skills/<name>/SKILL.md, agentskills.io format) for Blender 5.2 LTS+. Codex, OpenCode, Gemini CLI, Copilot and Cursor read them from .agents/skills; Claude Code reads them from .claude/skills (see install/).

Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,542 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository29 other files it gives its agents.
- 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-hair-groomingSkill
- 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 AGENTS.md into the rules it gives an agent

5 rules. About 0 are common, 0 are distinctive, and 5 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Very specific:
Start any Blender task with skills/blender-core/SKILL.md.
line 6
Very specific:
Run scripts; don't read them.
line 7
Very specific:
Always finish geometry work with skills/blender-asset-qa (numbers + a visual check).
line 9
Very specific:
Test: python tools/run_tests.py [filter] (needs Blender 5.2.
line 45
Very specific:
Keep each SKILL.md ≤ ~2.5k tokens and each description ≤ 400 chars.
line 46

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

AGENTS.md

Blender Agent Skills: agent instructions

This repo contains Agent Skills (skills/<name>/SKILL.md, agentskills.io format) for Blender 5.2 LTS+. Codex, OpenCode, Gemini CLI, Copilot and Cursor read them from .agents/skills; Claude Code reads them from .claude/skills (see install/).

Using the skills

Start any Blender task with skills/blender-core/SKILL.md. It routes to the specialized skill and defines how to run code: live through the blender-mcp execute_blender_code, or headless with python skills/blender-core/scripts/run_in_blender.py <script.py> -- <args>.
Run scripts; don't read them. Every skills/*/scripts/*.py documents its flags in its docstring and prints one compact JSON result.
Load a skill's references/*.md only when the SKILL.md points you there.
Always finish geometry work with skills/blender-asset-qa (numbers + a visual check).

Skill	Use for
blender-core	execution, 5.x API rules, routing (start here)
blender-scene-setup	units, naming, collections, linking, asset libraries
blender-hardsurface-modeling	bevels, booleans, weighted normals, props
blender-sculpting	organic forms, multires, procedural sculpt ops
blender-retopology-uv	retopo, UVs, texel density
blender-pbr-materials	Principled BSDF, texture sets, color spaces
blender-texturing	wear/grime/dust layers, decals, trims, painting
blender-texture-baking	normal/AO/curvature/ID bakes, ORM
blender-geometry-nodes	procedural modeling, JSON node-tree builder
blender-character-rigging	Rigify, skinning, game skeletons
blender-facial-shapekeys	ARKit 52, visemes, correctives
blender-animation	actions/slots, NLA, mocap retarget, loops, bake, root motion
blender-clothing	garments, weights, body masking, morphs, poke check
blender-hair-grooming	hair curves, node assets, hair cards
blender-game-optimization	budgets, LODs, draw calls
blender-export-pipelines	Roblox, Unity, Unreal, Godot, web, USD
blender-rendering	Cycles/EEVEE, color management, batch
blender-lighting-lookdev	lights, HDRI, cameras, turntables
blender-stylized-npr	toon shading, outlines, Line Art, Grease Pencil
blender-environment-kits	terrain, scatter, modular kit checks
blender-video-editing	VSE edits, motion tracking
blender-simulation	rigid body, cloth (classic + 5.2 nodes), fluids
blender-time-of-day	sky/sun/moon by time, night, fog, moods
blender-weather-fx	rain, snow, wetness, puddles, lightning
blender-addon-development	extensions, operators, UI
blender-pipeline-automation	headless batch, conversion, CI
blender-asset-qa	review gate: z-fighting, floating parts, normals, UVs, weights
blender-workflow-recipes	end-to-end playbooks

Working on this repo

Shared helper: edit shared/_bpyx.py only, then run python tools/sync_shared.py.
Validate: python tools/validate_skills.py (spec, token budgets, links, script contract).
Test: python tools/run_tests.py [filter] (needs Blender 5.2; set BLENDER_EXE if it isn't found). Set BAS_SKIP_GPU=1 on machines without a GPU.
Keep each SKILL.md ≤ ~2.5k tokens and each description ≤ 400 chars. Verify every API claim against a running Blender 5.2 before writing it.

Download now (ZIP)

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

What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free