Figma Context MCP
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
作者による説明(英語の原文)
The Figma desktop app as an MCP server: full Plugin API for AI agents. Local, no token, no quota.
この3つを1つのスコアにまとめることはしません。意見、測定、人気はそれぞれ別のものだからです。
このサーバーのやさしい解説はまだ書いていません。以下はレジストリから読み取った事実です。
使っているAIアプリを選んで、設定をコピーしてください。
必要なもの
どのアプリを使っていますか?
上でアプリを選んでください。選んだアプリを覚えておくので、次からはボタンに「…にインストール」と表示されます。
Claude Desktop → Settings → Developer → Edit Config を開き、claude_desktop_config.json にこれを貼り付けます
{
"mcpServers": {
"localfig": {
"command": "npx",
"args": [
"-y",
"localfig"
]
}
}
}ターミナルでこれを実行します
claude mcp add localfig -- npx -y localfigプロジェクトの .cursor/mcp.json、または ~/.cursor/mcp.json に貼り付けます
{
"mcpServers": {
"localfig": {
"command": "npx",
"args": [
"-y",
"localfig"
]
}
}
}ブラウザがアプリを開いてよいか確認し、そのあとアプリが設定を表示して確認を求めます。キーの値は空欄のままなので、自分で入力してください。
プロジェクトの .vscode/mcp.json に貼り付けます
{
"servers": {
"localfig": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"localfig"
]
}
}
}ブラウザがアプリを開いてよいか確認し、そのあとアプリが設定を表示して確認を求めます。キーの値は空欄のままなので、自分で入力してください。
これはパソコンで動くサーバーですが、ChatGPTがつなげるのはインターネット上のサーバーだけです。ほかのアプリを選んでください。
選んだアプリは、この端末のCookieに保存されます。アカウントは不要です。
この設定は、作者のレジストリ登録情報から作成しています。READMEと違う場合は、READMEに従ってください。
パソコンで動くサーバーです。インストールすると、ツール一覧がAIアプリに表示されます。
誰でも見られる情報をチェックしており、コードは読んでいません。当てはまらない項目は数に入れていません。 · 2026年11月9日にチェック
レジストリ名は、リポジトリを所有するGitHubアカウント LucasGorgal で確認済みです
このリポジトリはまだGitHubから読み取っていません
このリポジトリはまだGitHubから読み取っていません
これはパソコンで動くため、インターネット上のエンドポイントはありません
パソコンで動き、代わりにローカルの設定を使います
パソコンで動くサーバーです。ツール一覧を読み取れるのは、インターネット上のエンドポイントがある場合だけです
レジストリでこのカテゴリにある有効なすべてのサーバーと比較しています。スターは、公開者がリポジトリの所有者である場合だけ数えます。
当てはまる基本チェック1項目のうち1項目に合格しました。コードは読んでいません。キーやトークンを追加する前に、サーバーがどんなアクセスを求めているかを確認し、必要な権限だけを持つキーを使ってください。
レジストリには必須のキーは載っていません。ただし、最初に使うときにサインインを求めるサーバーもあります。
あなたのパソコンで動きます。必要なときにAIアプリが起動し、データは作者のサーバーを通りません(サーバー自体がオンラインサービスを呼び出す場合を除きます)。
GitHubリポジトリの最終更新は–、最新バージョン(0.4.1)がレジストリに届いたのは2 か月前です。
Claude Desktop、Claude Code、Cursor、VS CodeなどのMCP対応アプリで使えます。パソコンにNode.jsも必要です。
同じカテゴリには:Figma Context MCP, Claude Real Video, Figma MCP Server
同じカテゴリで、ほかの作者が作ったサーバー。スターの多い順。
カードの見方:↓ 週あたりのダウンロードi★ GitHubスターi3つの柱:レビュー · チェック · 利用状況i更新=最新バージョンをレジストリに公開した日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.
作者のリポジトリから書かれたとおりにコピーしたものです。PopNx MCPは内容を確認・推奨していません。HTML、画像、スクリプトは取り除いています。リンクはこのサイトの外へ移動します。
The Figma desktop app as an MCP server. localfig gives an AI agent the full Figma Plugin API on the file you have open — through a tiny dev plugin and a loopback HTTP bridge, entirely on your machine.
MCP client --stdio JSON-RPC--> localfig --http localhost:8765--> Figma plugin --> Plugin API
| localfig | Official Figma MCP | REST-API servers | Fixed-command plugin bridges | |
|---|---|---|---|---|
| Writes to the file | anything the Plugin API can | limited | no — the REST API cannot write nodes | a fixed menu of commands |
| Arbitrary Plugin API JavaScript | yes (figma_eval) | no | no | no |
| Quota / login | none | plan-dependent | token + rate limits | none |
| Tokens, styles, components read back | figma_tokens, figma_metadata | yes, incl. Code Connect | partial | rarely |
| Agent can see the design as an image | yes, inline in the response | yes | files or links, not inline | varies |
| Undo per agent step, change tracking | yes | no | no | no |
npx -y localfig --setup
That generates your local secret, writes the plugin to ~/.localfig/plugin/, registers localfig with every MCP client it finds on your machine, and prints the manifest path for the next step. --dry-run shows what it would do without touching anything, and --clients=cursor,vscode limits it to the clients you name. Prefer a clone? git clone https://github.com/LucasGorgal/localfig.git && cd localfig && node server.mjs --setup does the same.
| Client | Where setup registers localfig |
|---|---|
| Claude Code | claude mcp add --scope user |
| Claude Desktop | claude_desktop_config.json |
| Cursor | ~/.cursor/mcp.json |
| Windsurf | ~/.codeium/windsurf/mcp_config.json |
| VS Code (Copilot agent mode) | user mcp.json |
| Cline (extension and CLI) | ~/.cline/data/settings/cline_mcp_settings.json, or VS Code's extension storage on older versions |
| Gemini CLI | ~/.gemini/settings.json |
| Codex | ~/.codex/config.toml |
| opencode | ~/.config/opencode/opencode.json |
| Zed | settings.json, under context_servers |
| LM Studio | ~/.lmstudio/mcp.json |
A client counts as installed when its config folder exists. Setup merges into existing files instead of replacing them: other servers and settings stay, an existing localfig entry keeps its env, and a .localfig.bak copy is written before any change. Files it cannot safely rewrite, such as a settings.json with comments, are left untouched and reported with the entry to paste. Restart any client that was already open.
For a client that is not in the list, add this to its MCP configuration:
{ "mcpServers": { "localfig": { "command": "npx", "args": ["-y", "localfig"] } } }
On Windows, use "command": "cmd", "args": ["/c", "npx", "-y", "localfig"] instead: most clients start servers without a shell, and npx is a .cmd script there.
Then import the plugin into Figma (once):
Ctrl / (Cmd /) and type *import plugin from manifest*.~/.localfig/plugin/manifest.json (the setup output prints the full path).The manifest asks for two permissions — teamlibrary (library components and tokens) and currentuser (who is driving) — and registers a relaunch button.
Ctrl / → *localfig* — or Ctrl Alt P (Cmd Opt P) to re-run the last plugin, or the Reconnect localfig button in the properties panel of a file it has run in. A small panel appears and says Connected to localfig once the MCP server is up. Leave it open.Closing the panel (or the file) disconnects; re-run the plugin to reconnect. This is the one manual step: Figma provides no way to launch a plugin from outside.
| Tool | What it does |
|---|---|
figma_status | Plugin connected? File, page, selection, top-level frames, who is driving, and whether dynamic code execution is available |
figma_eval | Run arbitrary Plugin API JavaScript in the file — the workhorse |
figma_metadata | Structural dump of a subtree: ids, names, types, geometry, text. styles: true adds fills (hex), gradient stops, fonts, effects, auto-layout, bound variables, named styles and component info per node; css: true adds the CSS Figma computes per node |
figma_tokens | Every local variable collection with its modes and per-mode values, plus paint / text / effect styles |
figma_find | Search the current page or every page by type, layer-name regex and text regex |
figma_changes | What changed since your last call — creations, deletions, property changes, per-node summary; edits made by tool calls are tagged and hidden by default, so this is what the person did in Figma meanwhile |
figma_history | snapshot: save a named version (restore point). undo: revert the last tool call — every mutating call is checkpointed as one undo step |
figma_library | Team library: list variable collections and their variables, import components / styles / variables by key, place instances |
figma_export | Export nodes to PNG/JPG/SVG/PDF/JSON on disk. PNG/JPG come back inline as images; JSON (the subtree in Figma REST API shape) and SVG come back inline as text |
figma_place_image | Put a local image into a node's fill. Any format the browser decodes — PNG, JPG, WebP, GIF, BMP, AVIF. Images over 4096px per side are downscaled automatically (maxSide) |
figma_eval environmentCode runs inside the plugin sandbox, wrapped in an async function: top-level await and return both work.
const n = await figma.getNodeByIdAsync('10:59'); // sync getNodeById is unavailable (dynamic-page access)
await helpers.setText(n.findOne(x => x.type === 'TEXT'), 'new copy'); helpers.reveal(n); // put it on the person's screen return { id: n.id, w: n.width }; // JSON-serialized back to the agent
Globals: figma and helpers.
helpers.createText({characters, font: {family, style}, fontSize, color: '#hex', width, lineHeight, letterSpacing, textCase, textAlign, name, parent, x, y}) — loads the font and orders resize() before textAutoResize.helpers.setText(node, chars) — loads the node's real fonts, then sets characters. helpers.loadNodeFonts(node).helpers.set(node, props) — batch assign; layoutMode first, width/height via resize, sizing modes after.helpers.rgb('#rrggbb'), helpers.rgba('#rrggbbaa') — hex to the {r, g, b[, a]} fills want.helpers.reveal(nodes) — scroll and zoom the viewport to what you just made. helpers.notify(msg) — a toast.helpers.collection(name, [modes]), helpers.token(collection, name, type, value | {mode: value}), helpers.bind(node, 'fills' | 'strokes' | prop, variable) — create design tokens and bind properties to them, one line each.helpers.importComponent(key), helpers.instance(key, parent, props) — team-library components.helpers.query(root, sel) — minimal selectors: TEXT, [name=X], [name*=X], A B, a, b. helpers.createAutoLayout(dir, props).helpers.command(kind, payload) — run any typed command (tokens, metadata, find, changes, export…) from inside a script: (await helpers.command('tokens', { collection: 'Brand' })).result.resize() on a TEXT node resets textAutoResize to NONE; on an auto-layout frame it sets both sizing modes to FIXED. Resize first, then set the sizing.figma_eval. Real Figma nodes are collapsed to {id, name, type} so a stray node reference cannot flood the result — use figma_metadata to inspect a subtree.figma.notify() works here (the cloud MCP blocks it).figma_tokens returns the file's variables the way a design system thinks about them:
{ "collections": [{ "name": "Brand", "modes": ["Light", "Dark"], "defaultMode": "Light",
"variables": [{ "name": "color/bg", "type": "COLOR", "values": { "Light": "#ffffff", "Dark": "#04060c" } }, { "name": "space/md", "type": "FLOAT", "values": { "Light": 16, "Dark": 16 } }, { "name": "color/accent", "type": "COLOR", "values": { "Light": "{color/blue-500}", "Dark": "{color/blue-300}" } }] }], "styles": { "paint": [...], "text": [...], "effect": [...] } }
figma_metadata with styles: true then reports, per node, boundVariables (which token drives each property), fillStyle / textStyle / effectStyle names, and for instances the main component and its componentProperties; css: true adds the CSS Figma computes for each node. Together they let an agent build on a design system instead of hard-coding values — and hand code off from a real file.
figma_place_image sends the file's bytes to the plugin UI, which is a full Chromium iframe. It decodes any browser-supported format, downscales to maxSide (default 4096, Figma's hard limit) with high-quality smoothing, converts non-native formats to PNG, and hands the result to the main thread. The tool result reports what happened (prepared).figma_export writes files to ~/.localfig/exports/ and attaches PNG/JPG (≤ 2 MB each) to the result as images, so the agent sees the render without another call. JSON exports the subtree in Figma REST API shape.Every mutating tool call is checkpointed with figma.commitUndo, so it lands in the file's undo history as one step a person can Ctrl+Z (the most recent call is committed 60 s after it finishes, or at the next call). Within that window figma_history {action: "undo"} reverts the last call from the agent side — triggerUndo reverts to the last checkpoint — and {action: "snapshot"} saves a named version before risky edits. The plugin also records node changes on every page it has seen (Figma delivers them batched, after the fact): figma_changes returns what changed since a seq, with edits caused by tool calls tagged byPlugin by a time-window heuristic and hidden by default — the agent can notice what the person changed in Figma between calls instead of overwriting it.
The first localfig process owns the bridge port; later ones detect it via /health, run as clients, and proxy through it — same tools either way. If the owner dies, the next tool call on a client re-binds the port and takes over ("re-election" in the log); the plugin re-registers when it reconnects.
~/.localfig/
token local secret (generated on first run) plugin/ manifest.json + code.js + ui.html — import THIS manifest into Figma e
冒頭だけを表示しています。続きはリポジトリで読めます