AI 에이전트가 Blender 워크플로우 내에서 클라우드 렌더링, 3D 모델링, 에셋 검색 및 캐릭터 애니메이션을 직접 수행할 수 있도록 지원합니다.
다른 이름 loopcut-ai/loopcut
- 스타
- 3
- Tools
- —
- 다운로드, 30일
- 933
- 최신 버전
- 0.3.0
- 마지막 커밋
- 2026년 11월 6일
- 라이선스
- GPL-3.0
Loopcut MCP란?
Loopcut은(는) MCP server입니다. AI 에이전트가 Blender 워크플로우 내에서 클라우드 렌더링, 3D 모델링, 에셋 검색 및 캐릭터 애니메이션을 직접 수행할 수 있도록 지원합니다.
프로젝트 README에서 생성된 요약
Loopcut MCP server 정보
Loopcut은 AI 에이전트가 Blender 워크플로우 내에서 복잡한 3D 제작 작업을 직접 실행할 수 있는 가교 역할을 합니다. Claude Desktop, Cursor 또는 Claude Code와 같은 클라이언트를 사용하는 개발자는 RTX 4090 GPU를 이용한 클라우드 렌더링, 무료 에셋 라이브러리 검색, 이미지 또는 텍스트 기반 3D 모델 생성을 수행할 수 있습니다. 이 MCP server는 비용 견적 제공부터 최종 파일을 로컬 환경으로 다시 다운로드하는 것까지 창의적 작업의 전체 라이프사이클을 관리합니다.
사용자는 캐릭터 리깅을 자동화하고, 메쉬를 깨끗한 지오메트리로 리메쉬하며, 비디오나 텍스트 설명을 통해 애니메이션을 생성할 수 있습니다. 시스템은 씬을 패킹하고 필요한 에셋을 클라우드에 업로드하여 파일 관리를 처리합니다. 모든 작업은 Loopcut 계정과 연결된 크레딧 기반 시스템을 통해 관리됩니다. 많은 도구가 클라우드에서 실행되지만, 모션 적용과 같은 특정 기능은 사용자의 로컬 Blender 인스턴스를 사용하여 실행될 수 있습니다. 설정에는 uv가 필요하며, 브라우저 기반 로그인 또는 자동화 환경을 위한 API 키를 통해 구성할 수 있습니다.
활용 사례
- 1Blender 씬 클라우드 렌더링고성능 클라우드 GPU를 사용하여 .blend 파일을 렌더링하기 위한 가격 견적을 요청합니다.
- 2자동 캐릭터 리깅휴머노이드 캐릭터를 Mixamo 본 이름과 스킨 웨이트가 적용된 리깅 모델로 변환합니다.
- 3이미지 기반 3D 모델 생성프롬프트나 참조 이미지를 제공하여 텍스처가 포함된 3D 모델을 생성합니다.
- 4비디오 기반 애니메이션 생성MP4 또는 MOV 파일에서 전신 동작을 추출하여 디지털 리그에 적용합니다.
- 5무료 3D 에셋 검색Poly Haven 및 ambientCG와 같은 무료 라이브러리에서 재질이나 모델을 찾습니다.
프로젝트의 README를 바탕으로 작성되었습니다. 모든 활용 사례는 해당 내용의 인용문과 대조하여 확인합니다.
Loopcut MCP 설치 방법
형식 확인: 2026년 11월 10일Claude Code에 Loopcut MCP 설치
$ claude mcp add --env LOOPCUT_API_KEY=<YOUR_LOOPCUT_API_KEY> loopcut -- uvx loopcutCursor에 Loopcut MCP 설치
~/.cursor/mcp.json
{
"mcpServers": {
"loopcut": {
"command": "uvx",
"args": [
"loopcut"
],
"env": {
"LOOPCUT_API_KEY": "<YOUR_LOOPCUT_API_KEY>"
}
}
}
}VS Code에 Loopcut MCP 설치
.vscode/mcp.json
{
"servers": {
"loopcut": {
"type": "stdio",
"command": "uvx",
"args": [
"loopcut"
],
"env": {
"LOOPCUT_API_KEY": "<YOUR_LOOPCUT_API_KEY>"
}
}
}
}Claude Desktop에 Loopcut MCP 설치
claude_desktop_config.json
{
"mcpServers": {
"loopcut": {
"command": "uvx",
"args": [
"loopcut"
],
"env": {
"LOOPCUT_API_KEY": "<YOUR_LOOPCUT_API_KEY>"
}
}
}
}Codex에 Loopcut MCP 설치
~/.codex/config.toml
[mcp_servers.loopcut] command = "uvx" args = ["loopcut"] [mcp_servers.loopcut.env] LOOPCUT_API_KEY = "<YOUR_LOOPCUT_API_KEY>"
Gemini CLI에 Loopcut MCP 설치
~/.gemini/settings.json
{
"mcpServers": {
"loopcut": {
"command": "uvx",
"args": [
"loopcut"
],
"env": {
"LOOPCUT_API_KEY": "<YOUR_LOOPCUT_API_KEY>"
}
}
}
}구성
LOOPCUT_API_KEY비밀A Loopcut API key (lc_...) from loopcut.org/account. Optional: without it, the loopcut_login tool signs in with a browser link.
LOOPCUT_BLENDERPath to the Blender executable used to pack scenes. Optional: found on PATH and in the usual install folders.
LOOPCUT_TEST_BLENDER
| 이름 | 설명 | 유형 |
|---|---|---|
| LOOPCUT_API_KEY | A Loopcut API key (lc_...) from loopcut.org/account. Optional: without it, the loopcut_login tool signs in with a browser link. | 비밀 |
| LOOPCUT_BLENDER | Path to the Blender executable used to pack scenes. Optional: found on PATH and in the usual install folders. | |
| LOOPCUT_TEST_BLENDER | — |
자동 검사
2026년 11월 13일에 실행됨전체 보고서 →자동 검사를 통과했습니다. 이는 보안 감사가 아닙니다. 코드를 직접 읽고 서버에 필요한 최소한의 권한만 부여하십시오. 검사 방식
- 소스 저장소 연결 및 접근 가능loopcut-ai/loopcut is public and reachablegithub.com/loopcut-ai/loopcut ↗통과
- 패키지가 이 저장소를 소스로 지정loopcut names loopcut-ai/loopcut as sourcepypi.org/project/loopcut/ ↗통과
- 빌드 출처(Sigstore 증명)Sigstore attestation for loopcutpypi.org/project/loopcut/ ↗통과
- 오픈 소스 라이선스 (OSI)GPL-3.0spdx.org/licenses/GPL-3.0.html ↗통과
- 최근 1년 내 커밋 또는 릴리스Last commit or release 2026-11-06github.com/loopcut-ai/loopcut ↗통과
- OpenSSF Scorecard검사되지 않음
- 패키지 이름이 인기 패키지와 유사하지 않음No popular package one edit away from loopcut통과
- 패키지와 의존성에 심각한 보안 권고 없음1 resolved dependencies checked; no critical advisoryosv.dev ↗통과
- 모델에 대한 숨겨진 지침 없음0 tool descriptions and the README read; no override phrases, credential paths or hidden characters통과
- 파괴적인 도구에 라벨 표시됨검사되지 않음
- 원격 endpoint가 MCP 핸드셰이크에 응답검사되지 않음
README
GitHub에서 보기loopcut
The connector for Loopcut, an open-source AI agent inside Blender: cloud render,
assets, images, 3D models, motion and audio for Claude Code, Codex, Cursor, Claude Desktop and the shell, and
as a hosted connector for claude.ai and ChatGPT. Render .blend files on cloud RTX 4090s, search free asset
libraries, make and edit images, turn images into 3D models, remesh meshes, rig characters, make sound effects and voice
lines, and turn a video or a sentence into character animation on your own rig. Results come back as files the agent imports with Blender's Python. Everything is paid from your
Loopcut account's credits. You don't need the Loopcut app; any Blender install packs the scene.
It's one MCP server plus a CLI with the same verbs. The tools come from Loopcut's catalog, so a cloud tool that opens on Loopcut shows up without a new release.
| Tool | What it does | Result |
|---|---|---|
loopcut_login |
Signs in or switches account. Any tool does it by itself the first time: a browser opens, you approve, and the call carries on | |
loopcut_account |
Plan, credit balance, Private mode and no-generation settings, and which tools you can use | |
render_quote |
Packs the saved scene with headless Blender, uploads what the cloud lacks and prices the render | PNG frames and a QA report |
asset_search |
Searches Poly Haven, ambientCG and Physically Based. Free, and it runs at once | Matches with licence and download details |
image_generate |
Makes an image from a prompt, optionally guided by a reference image: concept art, backplates, textures (transparent background on request) | PNG |
image_edit |
Changes an existing image from an instruction, with an optional mask: a paint-over or restyle of a render | PNG |
image_to_3d |
Makes a textured model from a prompt, an image or both | GLB |
remesh |
Rebuilds a GLB mesh with clean quads or triangles and bakes its textures onto it | FBX (quads) or GLB (triangles) |
auto_rig |
Rigs a humanoid character in a saved .blend: a skeleton with Mixamo bone names and skin weights on the original meshes. Your Blender checks it first for free; the finished job binds the rig and saves the file |
the rigged .blend |
sound_effect |
Makes a sound effect from a description | WAV |
voice_line |
Speaks a line with a chosen voice and direction | WAV |
job_start |
Starts a quoted job, which spends credits. Calling it again returns the same job | |
job_status |
Waits for a job, or up to 10 at once (job_ids), and downloads the results, hash-checked and resumable. Running jobs say about how long is left |
|
job_cancel |
Cancels a job and refunds unspent credits |
Results are files the next tool takes: a generated image as image_to_3d's image, a model as remesh's
model, a rendered frame as image_edit's image. With a budget from you, the agent passes max_credits to a
paid tool, which starts a quote within it at once instead of returning it for job_start.
A character from a photo, rigged and animated: image_edit the photo into a T-pose with separated fingers,
image_to_3d from that with topology: "quads", import it and save the .blend, auto_rig it (20 credits),
then motion_from_text and motion_apply. auto_rig takes humanoids with clear arms, legs and fingers, up to
100,000 triangles; a rig that can't be bound is refunded automatically.
Motion
Two motion tools, with a free local tool to use their results:
| Tool | What it does | Price | Result |
|---|---|---|---|
motion_from_video |
Captures a person's whole-body motion, hands included, from a local MP4, MOV or WebM (NVIDIA GEM-X). One person, whole body in view, a mostly still camera; at most 60 s per job, chosen with start_s and end_s, from a file up to 200 MB |
3 credits per second captured, 2-second minimum | motion.bvh + motion.json |
motion_from_text |
Makes 2–10 s of body animation from a description of one simple physical action (NVIDIA Kimodo, Built with Meta Llama 3). Good at grounded motion (walks, dances, kicks, gestures, fighting in place); flips, spins in the air and big jumps come out small or missing. The fingers stay still; it isn't for faces, props or finger work | 2 credits per second of motion | motion.bvh + motion.json |
motion_apply |
Retargets a result onto the humanoid rig in a saved .blend with your Blender, headless, and saves the file. Mixamo, Rigify, Auto-Rig Pro, Unreal mannequin, or any humanoid by bone names; with no rig, SOMA's own skeleton with a simple body. Free and local |
free | the animated .blend |
motion.bvh is a rig-neutral SOMA 77-joint skeleton (standing T-pose rest, centimetres, Y up, 30 fps).
The connector measures the clip's length itself (MP4 and MOV from the file header, WebM with ffprobe
when it's installed) because the quote holds the seconds captured. target_armature and in_place on a
motion tool are kept on your machine for motion_apply and never sent to Loopcut. job_status returns
the result with three ways to use it: motion_apply (or loopcut motion-apply) for a saved .blend,
a short Python snippet for a Blender that's already open (for example through another Blender MCP),
and Blender's plain BVH import. The retargeting is the Loopcut add-on's own, so results match the app.
With no-generation on, motion_from_video stays available (it records the user's own performance) and
motion_from_text is hidden.
Price. 1 credit = $0.01. Every paid tool quotes first, for free, and the quote is the most a job can cost. Rendering costs 5 credits per GPU-minute on an RTX 4090 with a one-minute minimum; the other prices are on loopcut.org/pricing. Asset search, quotes, packing and the render QA report are free. Failed jobs are refunded. New accounts start with 100 credits.
Install
You need uv (it provides Python 3.10+), and Blender if you render or apply a
motion. To get uv: brew install uv or curl -LsSf https://astral.sh/uv/install.sh | sh on macOS and Linux,
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex" on Windows.
claude.ai, ChatGPT and Claude Desktop, nothing to install: add https://loopcut.org/mcp as a connector and
sign in with Loopcut. In Claude: Customize → Connectors → Add custom connector. In ChatGPT: chatgpt.com/plugins →
+ → Add custom MCP server (custom servers with paid tools need a Business, Enterprise or Edu workspace
until Loopcut is in ChatGPT's plugin directory). The hosted connector has asset search, images, 3D models,
remeshing, motion from text, sound, voice and rendering from a script, and returns download links; one tool's
result feeds the next by its job:<job_id>/<name> ref. Rendering your own .blend and motion from a video
need the plugin below, because they read files on your computer.
Claude Code (plugin, which includes a skill that teaches the agent the tools and the habits that save credits)
/plugin install loopcut --marketplace loopcut-ai/loopcut
On Claude Code before 2.1.275, run /plugin marketplace add loopcut-ai/loopcut and then
/plugin install loopcut@loopcut. Or add only the MCP server:
claude mcp add loopcut -- uvx loopcut mcp
Codex (plugin, with the same skill)
codex plugin marketplace add loopcut-ai/loopcut
codex plugin add loopcut@loopcut
Or add only the MCP server:
codex mcp add loopcut -- uvx loopcut mcp
Every tool call returns within 50 s, under Codex's default 60 s tool timeout; long work carries on between
calls. Both plugins start the server with uvx loopcut==<release> mcp, pinned to the release they ship with; the
package has no dependencies beyond Python.
Cursor, VS Code, Claude Desktop and other MCP clients
{
"mcpServers": {
"loopcut": {
"command": "uvx",
"args": ["loopcut", "mcp"]
}
}
}
Apps opened from the Dock or Start menu may not see your shell's PATH. If the server doesn't start, use
the full path from which uvx as command.
The first call opens a browser to sign in, and the key is kept where only you can read it. Without a
browser (over SSH, say), the call returns the link for you to open. LOOPCUT_API_KEY skips all of this.
CLI
loopcut login # once per machine
loopcut account # plan, credits, settings
loopcut tools # the open tools and their inputs
loopcut render shot.blend --frames 1-120 --samples 256 --max-credits 300
loopcut run asset_search --input '{"query": "oak floor", "kind": "material"}'
loopcut run image_to_3d --input '{"prompt": "a brass desk lamp", "quality": "draft"}' --file image=lamp.png --max-credits 30
loopcut quote sound_effect --input '{"prompt": "a heavy wooden door slams", "duration_seconds": 2}'
loopcut run motion_from_video --file video=take.mp4 --input '{"static_camera": true}' --blend shot.blend --max-credits 60
loopcut run motion_from_text --input '{"prompt": "a person jumps over a puddle", "duration_s": 4}' --yes
loopcut motion-apply --blend shot.blend --bvh loopcut_outputs/JOB_ID/motion.bvh --armature Hero --in-place
loopcut motion-apply --bvh loopcut_outputs/JOB_ID --bpy # Python for a Blender that's already open
loopcut start QUOTE_ID | status JOB_ID --wait | fetch JOB_ID --out DIR | cancel JOB_ID
loopcut settings --private-mode off # or on; also --no-generation on|off
Install the command with uv tool install loopcut, or run any command without installing as uvx loopcut ….
run and render quote, then start only with --yes, under --max-credits N, or after you answer the
prompt. run of a motion tool with --blend also puts the motion on that scene's rig; motion-apply
saves over --blend (Blender keeps the previous file as .blend1) unless you give --out. Results print as JSON on stdout; progress goes to stderr. In CI, create a key on
loopcut.org/account and set LOOPCUT_API_KEY.
Settings
| Variable | Meaning |
|---|---|
LOOPCUT_API_KEY |
A Loopcut key (lc_…) to use instead of the stored sign-in, e.g. in CI |
LOOPCUT_BLENDER |
The Blender executable used for packing and for motion_apply. Without it, the tool checks BLENDER, then PATH, then the usual install folders |
LOOPCUT_SERVICE_URL |
The service, https://loopcut.org by default |
LOOPCUT_CONNECTOR_HOME |
Where the sign-in, quotes, jobs and the last tool catalog are kept. Default: ~/.config/loopcut-connector, or %APPDATA%\loopcut-connector on Windows. It is separate from the Loopcut app's own files |
Private mode and no-generation. These account settings apply here too. Private mode (on by default on
paid plans) refuses tools whose provider may train on what it receives; today that is sound effects.
No-generation hides and refuses the generative tools. Change them in the Loopcut app or with
loopcut settings.
How it works
- Quote. Every tool goes through the same job API the Loopcut add-on uses. A quote prices the exact request and lasts ten minutes; an expired one is renewed at start, and the job goes ahead only if the price didn't change. Starting the same quote twice returns the same job.
- Files in. File inputs are local paths. They're checked (type and size), addressed by SHA-256, and only uploaded if the cloud doesn't already have them, straight to storage with presigned URLs. Big files go in parts, each retried on its own.
- Render packing. A background Blender (
-b --factory-startup --disable-autoexec) opens the saved.blendand runs the Loopcut add-on's own collector (addon/bundles.py, vendored). It copies every texture, library, cache and sequencer file into a bundle with relative paths. That Blender gets a private config folder and none of your environment's keys. The cloud never runs Python from your file, and the renderer has no internet access. - Files out. Results are checked against their hashes and saved into
loopcut_renders/<job_id>/next to the.blendfor a render, andloopcut_outputs/<job_id>/in the working folder for everything else (orout_dir). The working folder is the client's workspace root when it shares one; Codex starts plugin servers in the plugin's own folder, so there results go under your home folder instead. A render's QA report flags black, white, flat and transparent frames, fireflies, duplicate frames and exposure jumps. - Motion onto a rig.
motion_applystarts the same kind of background Blender on your saved.blendand runs the add-on's retargeter (addon/motion_retarget.py, vendored) throughloopcut/motion_blender.py: one whole-body rest alignment, limb-only direction alignment for A-pose rigs, the pelvis path scaled by leg length from the character's own standing height, and keys written parent first. A previous action on the rig is kept in a muted NLA track. Nothing leaves your machine and it costs nothing. - Rigging.
auto_rigruns the add-on's rigging code (addon/autorig.pyandrig_*.py, vendored) in a background Blender throughloopcut/rig_blender.py, once on each side of the cloud job: first it checks the named meshes and writes the upload (the meshes as a GLB and a rig task with their fingerprints) to the state folder, without changing the.blend; after the job it checks the character is unchanged, verifies the returned skeleton and weights against what was sent, binds them to the original meshes and saves.
If the catalog can't be read at startup, the server uses the last catalog it read from that service, then
the snapshot in src/loopcut/catalog.json.
What it runs and connects to. uv, which provides Python (and downloads it if needed); your Blender,
headless, only to pack a scene for render_quote, to apply a motion with motion_apply and to check and bind a rig for auto_rig; ffprobe, if
installed, only to measure a WebM clip; and your browser, only to open the Loopcut sign-in
page. It connects to loopcut.org (the catalog, sign-in and job API) and to the presigned storage URLs
that API hands out for uploads and downloads. It sends no analytics.
Privacy Policy
Loopcut's privacy policy is at loopcut.org/privacy. For this connector:
- What it sends. The Loopcut key on every request, with the client's name and version. For a render,
the packed scene: the saved
.blendand every texture, library and cache it uses. For other tools, the inputs you or the agent give (prompts, settings) and the files named as inputs (images, meshes, videos). It reads nothing else from your machine: no other files, no environment keys besidesLOOPCUT_API_KEY, no open Blender session.motion_applyopens and saves the.blendyou name on your machine only, andtarget_armatureandin_placestay local too. Forauto_rigit sends the named meshes (as a GLB) and their fingerprints, not the.blend, and binds the result on your machine. - Where it's stored. Loopcut's database (Supabase, US) holds your account, quotes and jobs; uploaded files and results are kept in Cloudflare R2.
- Third parties. Each job goes only to the provider that runs it: today OpenAI, fal and the model developers it runs (Deemos for Rodin, Tripo), Google Cloud, ElevenLabs and RunPod. Providers' own retention terms are on the privacy page. Private mode, on by default for paid plans, refuses tools whose provider may train on what it receives (today, sound effects).
- Retention. Uploaded files are deleted 7 days after their last use and results 30 days after theirs. Account and billing records are kept while the account exists. Results you download are ordinary files on your machine.
- On your machine. The key, quotes, jobs and the last tool catalog are kept in
~/.config/loopcut-connector, readable only by you.loopcut logoutremoves the key. - Contact. hello@loopcut.org, also for exporting or deleting your account data.
Development
PYTHONPATH=src python3 -m unittest discover -s tests # MCP protocol and unit tests
LOOPCUT_TEST_BLENDER=/path/to/blender PYTHONPATH=src python3 -m unittest tests.test_blender_pack
python3 scripts/snapshot_catalog.py # refresh the fallback catalog before a release
The end-to-end tests run against Loopcut's contract mock, which lives in the service repository; they're
skipped here. addon/VENDORED.md explains how the add-on copies are kept in sync.
Released under the GPL-3.0-or-later. Loopcut is an independent project; Blender is a trademark of the Blender Foundation.
Loopcut MCP 질문
Loopcut MCP를 Claude Code에 추가하려면 어떻게 하나요?
터미널에서 claude mcp add --env LOOPCUT_API_KEY=<YOUR_LOOPCUT_API_KEY> loopcut -- uvx loopcut를 실행합니다. 모든 프로젝트에서 사용할 수 있게 하려면 --scope user를 추가합니다.
Loopcut MCP를 Cursor에 추가하려면 어떻게 하나요?
~/.cursor/mcp.json를 열고 mcpServers 아래에 서버를 추가합니다. 위 설치 섹션에 표시된 Cursor 구성을 사용합니다.
Loopcut 사용 비용은 얼마인가요?
비용은 크레딧 기반으로 산정되며, 1 크레딧은 $0.01와 같습니다. 모든 유료 도구는 작업을 시작하기 전에 무료 견적을 제공합니다.
별도의 앱을 설치하지 않고도 Loopcut을 사용할 수 있나요?
네, 모든 Blender 설치 버전은 씬을 패킹할 수 있으므로 Loopcut 앱이 필요하지 않으며, Claude.ai 및 ChatGPT를 위한 호스트 커넥터로 작동합니다.
작업이 실패하면 어떻게 되나요?
작업이 실패하면 크레딧이 환불됩니다.
서버 인증은 어떻게 하나요?
첫 호출 시 브라우저가 열려 로그인을 진행하거나, CI와 같은 환경에서는 LOOPCUT_API_KEY를 사용할 수 있습니다.