Figma Context MCP
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
作者提供的说明(英文原文)
Local video editing for AI agents: cut, caption, analyse, reuse viral formats. FFmpeg, no cloud.
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"wecut": {
"command": "npx",
"args": [
"-y",
"wecut",
"mcp"
]
}
}
}在终端中运行这条命令
claude mcp add wecut -- npx -y wecut mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"wecut": {
"command": "npx",
"args": [
"-y",
"wecut",
"mcp"
]
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"wecut": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"wecut",
"mcp"
]
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日
注册名称已通过 GitHub 账户 13kailouis-stack 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于–,最新版本(0.1.0)于14天前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
同一分类中有:Figma Context MCP, Claude Real Video, Figma MCP Server
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
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.
The Figma MCP server brings Figma design context directly into your AI workflow.
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
WeCut is a free, open-source video editing engine for AI agents. The agent writes a small JSON project; wecut renders it with the FFmpeg already on the machine. It cuts, captions, reframes and mixes short-form video, and it can analyse a video that went viral and reuse its format as a template on new footage. CLI and MCP server. No cloud, no API key, no GPU, no telemetry.
Things you can ask an agent that has WeCut:
talk.mp4 and put the music under it, quieter when he speaks."npx -y wecut doctor # is FFmpeg ready?
npx -y wecut guide # the full guide for agents npx -y wecut probe media/talk.mp4 # duration, size, fps, audio npx -y wecut validate project.json # errors with JSON pointers and fix hints npx -y wecut preview project.json --at 1,5,9 # small JPG frames to look at npx -y wecut render project.json # out/clip.mp4 npx -y wecut mcp # the same, as an MCP server on stdio
Every command prints one JSON object on stdout. Exit codes: 0 ok, 2 invalid input, 3 missing dependency, 4 render failed, 5 over a limit.
WeCut Studio is an independent static web editor. Choose local clips, trim and reorder them, add text and music, then download a portable template for an agent on another device. Editing runs in the browser; rendering runs on the agent's computer. A public editor needs only static file hosting, not a rendering server. A localhost URL itself is not public.
Run python -m http.server 8770 --bind 127.0.0.1 --directory studio from this checkout and open http://127.0.0.1:8770. No build or backend is needed. The template file is shared through your agent chat; public template storage and project autosave are not included. Bundle import requires the 0.2.0 source/build; published npm 0.1.0 does not include it yet.
The quick way, for Claude Code, Claude Desktop, Codex, Cursor, Gemini CLI, Windsurf and Grok Build:
npx -y wecut setup # shows what it would change
npx -y wecut setup --apply # registers the MCP server and the skill, backs up each file it edits npx -y wecut setup --remove --apply
It only touches the clients it finds (or --client codex,cursor), keeps every other setting as it was, and reads each file back. Quit Claude Desktop before you run it: while it runs it can write its settings back and drop the entry. Run it again after updating WeCut, so the installed skill stays current. Or do it by hand:
WeCut needs FFmpeg 6 or newer with libass (npx -y wecut doctor checks it and says what to install):
sudo apt-get install -y ffmpegbrew install ffmpeg-full (the slim ffmpeg formula has no libass; WeCut finds ffmpeg-full on its own)winget install Gyan.FFmpegThen connect it to the agent you use. The MCP server works in a fixed workspace folder (--root, default: the folder it was started in); every media path is relative to it.
claude mcp add wecut -- npx -y wecut mcp
Or as a plugin, which also installs the WeCut skill:
/plugin marketplace add 13kailouis-stack/wecut
/plugin install wecut@wecut
Claude Desktop, Cursor, Windsurf, Cline, Roo Code, LM Studio and most other clients take this block in their MCP settings file. Give --root explicitly, because desktop apps start servers from an unpredictable folder.
{
"mcpServers": { "wecut": { "command": "npx", "args": ["-y", "wecut", "mcp", "--root", "/path/to/workspace"] } } }
code --add-mcp '{"name":"wecut","command":"npx","args":["-y","wecut","mcp"]}'
or in .vscode/mcp.json:
{ "servers": { "wecut": { "type": "stdio", "command": "npx", "args": ["-y", "wecut", "mcp"] } } }
codex mcp add wecut -- npx -y wecut mcp
or in ~/.codex/config.toml:
[mcp_servers.wecut]
command = "npx" args = ["-y", "wecut", "mcp"]
gemini extensions install https://github.com/13kailouis-stack/wecut
WeCut targets any agent with permission to run local programs and access the user's media. Product names do not determine compatibility: a terminal agent can use the CLI without MCP.
Grok Bot, Manus, Muse, OpenBot, Dots and Claude Duet are examples of requested integration targets, not a list of tested integrations. For each product, verify that its execution environment can run the WeCut binary and FFmpeg. A chat interface alone does not provide that access.
npx -y wecut guide # read this first: workflow, project format, templates
npx -y wecut doctor
Then follow the guide: probe, write project.json, validate, preview, render. In a fresh Linux sandbox install FFmpeg first with sudo apt-get install -y ffmpeg.
Clients that can launch a local MCP process can use wecut mcp over stdio; the client-specific examples above are not an allowlist. An HTTP-only MCP client needs a separate bridge: WeCut does not currently serve remote HTTP MCP. Browser-only agents need an execution environment before they can render. Node is needed for the npm launcher; a standalone binary does not need Node.
The skill in skills/wecut/SKILL.md teaches the workflow to agents that load skills (Claude Code, Claude apps, Codex, Cursor and others):
npx skills add 13kailouis-stack/wecut
npx -y wecut , or npm i -g wecut. npm installs only the binary for your platform (@wefluence/wecut- - ); there are no install scripts.SHA256SUMS.go build -o wecut ./cmd/wecutPrebuilt for Linux, macOS and Windows on x64 and arm64. Set WECUT_BINARY to use your own build through the npm launcher.
xfade), clip fadesgradients), like the dark fade behind captionsinset), colour washes for a stretch of the video, captions that light up word by word (reveal) with key words set apart (accent)--no-loudnorm skips it)wecut clone media/viral.mp4 --name news-card # -> recipes/news-card.json + out/clone/news-card/compare.jpg
clone turns a finished video into a template the way an editor would have built it: every cut and transition (flash, fade, dissolve, zoom, blur) at its exact time, each shot's framing (full screen, or shown whole on black or blurred bars) and push-in, the text cards with their colours, box, size, line spacing, padding, timing and entrance (fade, pop, slide or type), the caption style and position, the logo, and a talk that already plays under the opening B-roll. A steady beat is stored as beat indexes; recipe apply snaps those cuts onto replacement music. Each shot is a media slot, like a CapCut template. WeCut holds no model and no API key: it returns a review with the images and evidence behind each finding, and the agent running it looks and answers, so words, fonts, what each shot is and the kind of each transition come from real eyes and OCR is only a fallback. The comparison sheet puts the original beside the template filled with numbered stand-ins, so a person or a model can check it at a glance. People fill the slots in WeCut Studio; agents with wecut recipe apply. Details and the format: docs/clone.md. Cloning needs the 0.3.0 source/build; the published npm 0.1.0 does not have it yet.
wecut analyze media/viral.mp4 # shots, pace, hook, flashes, tempo, pauses + a frame sheet
wecut recipe list # templates: built-in and the workspace's recipes/*.json wecut recipe apply --input request.json # recipe + new clips, words, music -> project wecut recipe from-project my-edit.json --write recipes/my-style.json # "save as template"
analyze measures what a model cannot see in a transcript: where the cuts are (a white flash or a strobe counts once), how long the opening shot holds, whether cuts land on the beat, and whether the file ends with a silent platform outro. Its frame sheet, one frame per shot with timestamps, is how the agent reads the headline card, text style and framing.
A recipe (wecut.recipe/1, schema in schema/wecut.recipe.schema.json) is an edit's shape without its content: sections fed by roles (a 5 s hook with a headline card, then the body), text slots filled at apply time (with emphasis for the words to colour), caption and music style, effects on cuts. A recipe can carry its own default track, like a CapCut template; the music passed at apply time replaces it and "none" drops it. The built-in recipes copy the structure of formats that are common on Indonesian short-form video today: news-style reposts with a headline card, quote hooks, phrase-captioned vlogs, punched-in podcasts, photo slideshows.
Copy the structure, not the content: other creators' footage, wording and music stay theirs.
| Tool | Does |
|---|---|
wecut_doctor | checks FFmpeg, encoders and filters |
wecut_probe | duration, size, fps, audio of media files |
wecut_validate | checks a project; errors carry a JSON pointer and a hint |
wecut_preview | renders frames and returns them as images, so the model can look |
| wecut_render | renders M
只显示了开头部分;其余内容请在仓库中阅读