AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻译准备中
Headless Blender automation via FastMCP — 41 portmanteau tools (150+ ops). Batch mesh/VSE/Grease Pencil, VRM, Gaussian splats; optional live bridge. For Cursor / Claude Desktop. React dashboard + Taur
翻译准备中
开发工具Python · TypeScriptGitHub @sandraschi自行安装运营者未确认
试用过了吗? 点一下就好,无需登录。
个人 GitHub 账号下公开仓库的所有者可以登录并验证所有权,自行管理介绍,无需管理员批准。
登录并生成验证文件,在 30 分钟内将其提交到此仓库的默认分支。其他分支或分叉仓库中的文件无效。然后在下方检查文件,即可无需管理员批准直接管理项目。请保留文件以维持管理权限。
如需登录,将前往登录页面。登录后请回到此处,再次点击所选的验证按钮。
安装型项目无需部署域名也可验证。个人仓库可通过 GitHub 登录验证,组织仓库可使用上方的仓库验证文件。
Headless Blender automation via FastMCP — 41 portmanteau tools (150+ ops). Batch mesh/VSE/Grease Pencil, VRM, Gaussian splats; optional live bridge. For Cursor / Claude Desktop. React dashboard + Tauri app.
翻译准备中
取自仓库 README
正在加载完整 README。 以下为已保存的摘录。
Control Blender with natural language through MCP. Tell Claude to create a steampunk robot
with glowing eyes and watch it build in Blender.
1. Download the .mcpb from Releases
2. Drag it into Claude Desktop (Settings → MCP Servers → Install from file)
3. Install Blender if you have not already
4. Restart Claude Desktop and try: Create a red cube on a gray floor with a sun lamp.
That is enough for most tools — the server runs Blender headless in the background.
Optional — live viewport: install the bridge addon in Blender to watch the scene update while you chat. See INSTALL.md.查看完整 README ↗Control Blender with natural language through MCP. Tell Claude to create a steampunk robot with glowing eyes and watch it build in Blender.
| Download | What it is |
|---|---|
| blender-mcp-*.mcpb | MCP server for Claude Desktop |
| Blender | 3D app (install separately, never bundled) |
| blender_bridge_addon.py (optional) | Live viewport only — skip for headless use |
Claude Desktop one-liner (downloads + runs the installer from the latest release):
irm https://github.com/sandraschi/blender-mcp/releases/latest/download/install.ps1 | iex
Other MCP clients (manual config via uvx):
{
"mcpServers": {
"blender-mcp": {
"command": "uvx",
"args": ["blender-mcp"],
"env": { "BLENDER_EXECUTABLE": "C:\\Program Files\\Blender Foundation\\Blender 4.4\\blender.exe" }
}
}
}
Steps
.mcpb from ReleasesThat is enough for most tools — the server runs Blender headless in the background.
Optional — live viewport: install the bridge addon in Blender to watch the scene update while you chat. See INSTALL.md.
Other install methods (npx mcpb, manual config, developer setup): INSTALL.md
| Mode | Host app | When |
|---|---|---|
| Headless (default) | blender --background subprocess | Batch export, CI, agents without a display; VSE, geonodes, most mesh ops |
| Live GUI (optional) | Blender + bridge addon | Watch the agent build; viewport screenshots; sculpt with live feedback |
| Per-tool override | prefer_session=False | Batch jobs force headless even if bridge is connected |
You do not need to open Blender's UI for most MCP tools — the server spawns headless Blender automatically and falls back from live session when no bridge is connected.
Install Blender separately; it is never bundled. Override path with BLENDER_EXECUTABLE.
Watch or batch — Most tools run headless. Use
blender_sessionstart + the bridge addon only if you want the viewport to update live. See INSTALL.md.
| Direction | Artifacts | Notes |
|---|---|---|
| Hands-in | Natural-language scene prompts | Agent instructions; blender_ai_* script generation |
| Hands-in | .blend, image refs, mesh files | Webapp upload or tool params |
| Hands-in | Rodin / Tripo / Hunyuan mesh URLs | blender_ai_generate and related tools |
| Hands-in | Inline bpy scripts | blender_script_execute, handler-backed tools |
| Hands-out | .glb, .gltf, .fbx, .obj, .usd | blender_export — headless |
| Hands-out | .vrm, VRChat-ready avatars | Export + validation pipeline — headless |
| Hands-out | .blend (saved scene) | After agent edit session — headless or live bridge |
| Hands-out | Viewport PNG, MP4 (VSE), Gaussian splats | blender_render, blender_vse, splat tools — headless |
blender-mcp works completely standalone. You can optionally connect it with companion servers in the sandraschi fleet to unlock extended features:
| Companion Server | Feature Unlocked | Status | Setup |
|---|---|---|---|
qcad-mcp | dxf-svg-exchange, 2d-cad-to-3d-scene | Optional | Install Guide |
freecad-mcp | step-cad-import, mechanical-visualization | Optional | Install Guide |
resonite-mcp | gltf-export-to-vr, avatar-mesh-pipeline | Optional | Install Guide |
speech-mcp | task-completion-readout, render-completion-announcements | Optional | Install Guide |
Self-Contained Companions: Fleet companions operate independently. Installing companions does not trigger transitive dependency chains.
.mcpb packaging, Prometheus metrics, optional DockerTry these in Claude Desktop after install:
Create a red cube on a gray floor with a sun lamp and render a viewport screenshot.
Build a simple chair with wood material and export as GLB.
Start a live Blender session and add a sphere with a metallic blue shader.
| Doc | Contents |
|---|---|
| Installation | All install methods, prerequisites, bridge setup |
| Configuration | Env vars, Claude Desktop env block |
| Tool Reference | Capabilities and tool catalog |
| Development | Contributing, just, lint, build |
| Troubleshooting | Common errors and fixes |
| Architecture | System design |
| Roadmap | Planned improvements |
| Monitoring | Prometheus / Grafana / Loki |
| Docker | Optional container deploy |
Extended guides: docs/DOCUMENTATION_INDEX.md
Dashboard (optional): .\start.ps1 → http://localhost:10848 — see INSTALL.md.
Tauri desktop installer (~15 MB, no Python required): just build-native — see docs/DEVELOPMENT.md.
BLENDER_EXECUTABLEMIT — FlowEngineer sandraschi. Free for personal and commercial use.