Figma Context MCP
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
Description from the maker (original English)
Agent-native CAD: draw plans, model 3D, render checks, export DXF/SVG/STL/GLB/IFC, import DXF.
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json
{
"mcpServers": {
"modulor": {
"command": "uvx",
"args": [
"modulor"
]
}
}
}Run this in your terminal
claude mcp add modulor -- uvx modulorPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"modulor": {
"command": "uvx",
"args": [
"modulor"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"modulor": {
"type": "stdio",
"command": "uvx",
"args": [
"modulor"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
This one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Runs on your computer; the tool list appears in your AI app after you install it.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 6, 2026
The registry name was verified with the GitHub account bcllcc, which owns the repo
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
This one runs on your computer; there is no internet endpoint
Runs on your computer and uses local settings instead
Runs on your computer; we can only read tool lists from internet endpoints
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).
The GitHub repo last changed –, and the latest version (0.7.1) reached the registry 5 months ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Python (uv) on your computer.
In the same category: Figma Context MCP, Claude Real Video, Figma MCP Server
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
Let any LLM watch a video locally — and search everything it has ever watched.
Figma’s official server gives AI your design data so it can turn designs into code or describe them accurately. Connects to Figma’s cloud.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Agent-native 的二维绘图 + 三维建模内核。没有 GUI,没有鼠标——JSON 进,几何出。
English: Modulor is an agent-native 2D drafting + 3D modeling kernel —
the geometry layer for the agent era. No GUI: the entire tool is a
self-describing JSON op protocol (77 ops, contract-tested) over a plain-JSON
document format, callable via CLI, JSON-Lines pipe, MCP server or Python.
Parametric recipes, architectural semantics, freeform surfaces, DXF/glTF
interop, a built-in verification loop (measure / validate / render-and-see),
and a conformance checker for the format. Agents read
AGENT_GUIDE.md; the API contract lives in
docs/API.md; the file format spec in
docs/FORMAT.md. Named after Le Corbusier's *Modulor* —
a universal measure for building, this time for machines.
传统 CAD软件 是为人类的眼睛和手设计的:菜单、视口、 捕捉、快捷键。Agent 调用它们时,要么模拟人类操作,要么穿过厚重的插件层。 Modulor 反过来:把 CAD 的几何核心保留下来,把交互层整个换成 Agent 的母语 ——结构化命令、结构化结果、结构化错误,外加一条"渲染成图、亲眼验证"的反馈回路。
┌─────────────────────────────────────────┐
Agent ──────▶│ CLI batch │ CLI pipe │ MCP │ Python API │ └──────────────────┬──────────────────────┘ ▼ JSON 命令(自描述 op 协议) ▼ ┌─────────────────────────────────────────┐ │ 几何内核:manifold3d (Clipper2 + Manifold) │ │ 2D/3D 布尔 · 偏移 · 拉伸 · 旋转体 · 切片 │ └──────────────────┬──────────────────────┘ ▼ .json 文档 · SVG · DXF · PNG 渲染 · OBJ / STL / GLB
扩展生态:领域能力住扩展包——装上扩展即获得 mech.gear 渐开线齿轮(官方模板); 三十行代码就能写出你自己的扩展,命名空间与 API 法律由机制强制, 见 docs/PLUGINS.md。
身份与治理:项目名/包名/CLI/格式标识已正式冻结; Standard(规格,供第三方实现)与 Core(参考实现,canonical 而非排他)的 边界、semver 与兼容政策、RFC 流程见 GOVERNANCE.md。
标准的两半都已成文并由测试看守:op 协议——docs/API.md (由注册表生成)+ 机器可读契约 docs/api.json,实现与契约 不一致时测试直接失败;文档格式——docs/FORMAT.md 规格
每份示例文档都在 CI 里对着 schema 验证。
与主流 CAD 的互操作是契约(docs/INTEROP.md): DXF 导出为 R2000 原生对象(ELLIPSE/SPLINE/HATCH/关联 DIMENSION/ BLOCK+INSERT),在 AutoCAD 里直接可编辑;导入保留块结构;每个导出文件 在 CI 里由 ezdxf(DXF)与 ifcopenshell(IFC4)做语义级校验。
"bay*3"、"level_top('L2')"、"grid_x('B')");recipe_set 把生成 命令存进文档作为设计意图,set_param + regenerate 一条命令全模型 联动重建——"柱距 4m 改 5m,其他保持联动"就是这两个调用cad_render 直接返回图像modulor serve 起一个严格只读的浏览器窗口 (2D 平面 + WebGL 3D 轨道视图 + 图层开关 + 一键下载),监视文档文件, 任何 Agent 通过任何通道修改文档,页面自动跟随刷新——人类看,Agent 干pip install modulor # 内核依赖仅 numpy + manifold3d
pip install modulor[check] # 可选:格式一致性校验(modulor check --strict)
开发安装:克隆本仓库后 pip install -e .[dev],pytest tests 跑全部测试。
modulor ops # 看所有命令(自描述,无需读文档)
modulor ops add_wall # 看单个命令的参数/默认值/示例
echo [{"op":"add_box","size":[100,60,40]},{"op":"render","path":"a.png"}] > s.json modulor run model.json s.json # 批量执行(原子性:失败则不落盘)
modulor repl model.json # 长会话:一行 JSON 进,一行 JSON 出 modulor mcp # MCP stdio 服务器 modulor serve model.json # 只读查看器:浏览器实时跟随文档变化
*查看器:左 2D 平面(滚轮缩放/拖拽平移),右 WebGL 3D(轨道相机), 底部图层开关,顶部 SVG/DXF/GLB/STL 下载。零前端依赖,单文件 HTML,离线可用。*
MCP 客户端配置(如 Claude Code):
{"mcpServers": {"modulor": {"command": "modulor", "args": ["mcp"]}}}
接入你的 Agent:integrations/ 提供即插即用包——Claude Code skill、Cursor 规则、Codex AGENTS 片段、OpenAI 兼容工具定义(GPT/Kimi/通义/DeepSeek 直接加载)、LangChain/AutoGen/CrewAI 集成、中文接入说明、以及 8 个 CI 实测的命令配方(COOKBOOK)。
modulor run out\studio.json examples\floorplan.json —— 户型图: 闭合外墙环 + 隔墙 + 4 门窗 + 标注链 + 房间标签 → SVG/DXF/PNG → 一键长成 3D(含楼板)→ GLB。
modulor run out\bracket.json examples\bracket.json —— 机械零件: 偏移做圆角板 + 孔阵列 + 腰形槽(2D 布尔)→ 拉伸 → 凸台镗孔(3D 布尔)→ 体积测量 → STL/GLB/OBJ + 剖面 DXF。
modulor run out\zaha.json examples\zaha.json —— 异形建筑: 三断面样条放样出流线塔楼(loft+twist)+ 飘带天蓬(sweep)+ 样条曲线墙 + 隐式曲面融合展亭(add_implicit smin)→ GLB。
modulor run out\param.json examples\parametric.json --as-recipe —— 参数化建筑:5 个参数驱动轴网/标高/柱阵/楼板/幕墙/坡屋顶/楼梯/房间; 之后 {"op":"regenerate","params":{"bay":5000,"floors":4}} 一条命令, 整栋楼重新协调(OFFICE 面积 76.8m² → 120m²,diff 给出完整对比报告)。
| 建筑平面(双线墙、门扇、标注链) | 机械零件(2D/3D 布尔) | 异形建筑(放样/扫掠/隐式曲面) |
|---|---|---|
参数化联动——同一份配方,{"op":"regenerate","params":{"bay":5000,"floors":4}} 一条命令重建整楼:
| bay=4000 · 3 层 | bay=5000 · 4 层(regenerate 之后) |
|---|---|
| 城市体块(分区/裙房塔楼/扭转地标) |
|---|
规模参考(scripts/bench.py,20 层塔楼 401 条命令):构建 0.08s、 文档 0.04MB(墙体参数化存储)、着色渲染 2.2s、GLB 导出 0.05s。
help 一个调用拿全;参数拼错会收到 "did you mean ..." 。measure、validate、render——Agent 不必 盲信自己的输出。tag 让选择器稳定。modulor/
document.py 文档模型(图层/材质/实体/选择器/序列化) engine.py 批处理执行器(原子性、结构化错误) geometry.py 向量/仿射/弧离散化/墙体轮廓偏移 shapes.py 实体 <-> 几何内核(CrossSection/Manifold)桥 ops/ 全部操作:绘图、建模、参数化、建筑语义、变换、查询、导出 exporters/ SVG · DXF · OBJ · STL · GLB(全部手写,零重依赖) render/ 软件光栅器、笔画字体、2D/3D 渲染 cli.py run / op / repl / ops / new / info / serve / mcp mcp_server.py MCP stdio 服务器(手写 JSON-RPC,无 SDK 依赖) viewer/ 只读实时查看器(stdlib HTTP + 单文件 HTML + 自写 WebGL) tests/ 行为 + 契约 + 模糊测试(pytest) examples/ 户型图与机械零件全流程脚本 AGENT_GUIDE.md 给 Agent 看的完整使用指南
MIT
mcp-name: io.github.bcllcc/modulor