blender-forge
Skill0 starsChanged 18 days agoMITView on GitHub
Drive Blender through the blender-forge MCP server - connect, build scenes, review renders, save and export safely. Load before any task that touches Blender (modeling, materials, lighting, render, animation, import/export).
1 other project uses a version of this.
Adopted by
Most recently taken up by
- tdimino/claude-code-minoantheir own version, first seen Nov 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- tdimino/claude-code-minoan★ 41same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository4 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- blender-modelingSkill
- blender-render-reviewSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 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.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 11
Never assume the scene is empty: read scene_info first.
- Very specific: line 40
Every error has code, message and sometimes hint.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
blender-forge
Connect
forge_status. If it reportsBRIDGE_OFFLINE: callstart_headless(no window, good for renders and exports), or ask the user to open Blender; the extension starts the bridge with Blender.- Never assume the scene is empty: read
scene_infofirst. Usenew_scene(confirm=true)only when the user asked for a fresh scene.
Conventions
- Z is up, units are metres, rotations are degrees, colours are
#RRGGBB(sRGB). - Primitives are centred on their origin: a cube of size 2 spans -1..1. A 0.75 m high desk top sits at
z = 0.75; legs go at half their height. - Names are how you refer to things later; give every object and material a clear name.
Workflow
- Plan the pieces and real-world sizes (see
blender-modelingfor references). - Build in one go with
build_scene(validated first: nothing is created if the spec is wrong), or step by step withcreate_object,create_mesh,add_modifier,material,add_light,add_camera. - Look before you say it is done:
lookreturns a contact sheet (front, right, top, iso). Check proportions, floating or intersecting parts, missing pieces. Fix withtransformand look again. - Light and frame:
light_rigfor three-point light,frame_camerato fit the subject. - Final image:
render_settings(eevee for speed, cycles for quality), thenrender_still; check the returned image. - Deliver:
save(path=...)writes a copy;export_filefor .glb (web), .fbx (Unity/Unreal), .usd, .obj, .stl (3D printing).
Safety
deleteandnew_sceneneedconfirm=true; ask the user before removing their work.savenever overwrites the open file unlessoverwrite=true.run_pythonis disabled unless the user enabled it. Prefer the dedicated tools; if you need Python, say what the script will do first.- Assets from
search_assets/import_assetare Poly Haven CC0; the credit is stored on the objects.
Errors
Every error has code, message and sometimes hint. NOT_FOUND: check names with scene_info.
BAD_ARGUMENTS: the message names the field. CONFIRM_REQUIRED: ask the user, then retry with confirm.
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.

