Skip to main content
Glama
alucardeht

Figma MCP

by alucardeht

extract_assets

Extract all assets from a Figma frame, including icons, images, and composite groups, with automatic categorization and progress tracking.

Instructions

Extract all assets from a frame with progress tracking.

HOW IT WORKS:

  • Detects "composite groups" (image + decorative shapes) and exports them as single PNG

  • For composite groups, the ENTIRE group is exported as one image, preserving layout

  • Automatically categorizes into icons/, images/, and images/composites/

  • Uses smart naming based on component hierarchy

  • Shows progress: "Processing batch 1/5 - found 8 icons, 3 images, 2 composites"

  • Final summary with all file paths

  • Look for "isCompositeAsset: true" in the frame tree to identify composite groups

TYPICAL WORKFLOW:

  1. get_frame_info → see what assets exist and identify composite groups

  2. extract_assets → download all

  3. Check summary for file paths

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
file_keyYesFigma file key
page_nameYesPage name (partial match)
frame_nameYesFrame name (partial match)
output_dirNoOutput directory (default: ./figma-assets)./figma-assets

Implementation Reference

  • Core implementation: Extracts icons (SVG), images (PNG), and composite groups (PNG) from Figma frame. Batches API calls, smart categorization, saves to outputDir/icons|images|composites/, returns paths and summary.
    export async function extractAssets(ctx, fileKey, pageName, frameName, outputDir) {
      const { chunker, figmaClient } = ctx;
    
      const file = await figmaClient.getFile(fileKey, 2);
      const page = figmaClient.findPageByName(file, pageName);
      if (!page) throw new Error(`Page "${pageName}" not found`);
    
      const frameRef = figmaClient.findFrameByName(page, frameName);
      if (!frameRef) throw new Error(`Frame "${frameName}" not found`);
    
      const frame = await figmaClient.getNode(fileKey, frameRef.id);
      const assets = findAssets(frame, { collectBounds: true });
    
      const iconsDir = join(outputDir, "icons");
      const imagesDir = join(outputDir, "images");
      const compositeDir = join(imagesDir, "composites");
      await mkdir(iconsDir, { recursive: true });
      await mkdir(imagesDir, { recursive: true });
      await mkdir(compositeDir, { recursive: true });
    
      const results = { icons: [], images: [], composites: [], failed: [] };
      const assetMap = {};
      const batchSize = 10;
    
      for (let i = 0; i < assets.length; i += batchSize) {
        const batch = assets.slice(i, i + batchSize);
        const ids = batch.map((a) => a.id).join(",");
    
        try {
          const svgData = await figmaClient.getImage(fileKey, ids, "svg");
          const pngData = await figmaClient.getImage(fileKey, ids, "png", 2);
    
          for (const asset of batch) {
            try {
              if (asset.isComposite) {
                if (pngData.images[asset.id]) {
                  const pngResponse = await axios.get(pngData.images[asset.id], { responseType: "arraybuffer" });
                  const filePath = join(compositeDir, `${asset.name}.png`);
                  await writeFile(filePath, Buffer.from(pngResponse.data));
                  results.composites.push({
                    path: filePath,
                    uniqueName: asset.name,
                    originalName: asset.originalName,
                    section: getSectionFromPath(asset.path),
                    bounds: asset.bounds,
                    isComposite: true,
                  });
                  assetMap[asset.name] = filePath;
                }
              } else if (asset.category === "icon" && svgData.images[asset.id]) {
                const svgResponse = await axios.get(svgData.images[asset.id]);
                const filePath = join(iconsDir, `${asset.name}.svg`);
                await writeFile(filePath, svgResponse.data);
                results.icons.push({
                  path: filePath,
                  uniqueName: asset.name,
                  originalName: asset.originalName,
                  section: getSectionFromPath(asset.path),
                  bounds: asset.bounds,
                });
                assetMap[asset.name] = filePath;
              } else if (pngData.images[asset.id]) {
                const pngResponse = await axios.get(pngData.images[asset.id], { responseType: "arraybuffer" });
                const filePath = join(imagesDir, `${asset.name}.png`);
                await writeFile(filePath, Buffer.from(pngResponse.data));
                results.images.push({
                  path: filePath,
                  uniqueName: asset.name,
                  originalName: asset.originalName,
                  section: getSectionFromPath(asset.path),
                  bounds: asset.bounds,
                });
                assetMap[asset.name] = filePath;
              }
            } catch (err) {
              results.failed.push({
                name: asset.name,
                originalName: asset.originalName,
                error: err.message,
              });
            }
          }
        } catch (err) {
          batch.forEach((a) => results.failed.push({
            name: a.name,
            originalName: a.originalName,
            error: err.message,
          }));
        }
      }
    
      const response = chunker.wrapResponse(
        {
          frame: frame.name,
          outputDir,
          summary: {
            icons: results.icons.length,
            images: results.images.length,
            composites: results.composites.length,
            failed: results.failed.length,
          },
          icons: results.icons,
          images: results.images,
          composites: results.composites,
          assetMap,
          failed: results.failed,
        },
        {
          step: "Asset extraction complete",
          progress: `${results.icons.length} icons, ${results.images.length} images, ${results.composites.length} composite groups`,
          nextStep: "Assets saved to disk. Use extract_styles for design tokens.",
        }
      );
    
      return { content: [{ type: "text", text: JSON.stringify(response, null, 2) }] };
    }
  • Tool schema defining name, description, and input parameters (file_key, page_name, frame_name, output_dir). Part of toolSchemas array exported for MCP server.
        name: "extract_assets",
        description: `Extract all assets from a frame with progress tracking.
    
    HOW IT WORKS:
    - Detects "composite groups" (image + decorative shapes) and exports them as single PNG
    - For composite groups, the ENTIRE group is exported as one image, preserving layout
    - Automatically categorizes into icons/, images/, and images/composites/
    - Uses smart naming based on component hierarchy
    - Shows progress: "Processing batch 1/5 - found 8 icons, 3 images, 2 composites"
    - Final summary with all file paths
    - Look for "isCompositeAsset: true" in the frame tree to identify composite groups
    
    TYPICAL WORKFLOW:
    1. get_frame_info → see what assets exist and identify composite groups
    2. extract_assets → download all
    3. Check summary for file paths`,
        inputSchema: {
          type: "object",
          properties: {
            file_key: { type: "string", description: "Figma file key" },
            page_name: { type: "string", description: "Page name (partial match)" },
            frame_name: { type: "string", description: "Frame name (partial match)" },
            output_dir: { type: "string", description: "Output directory (default: ./figma-assets)", default: "./figma-assets" },
          },
          required: ["file_key", "page_name", "frame_name"],
        },
      },
  • src/index.js:60-62 (registration)
    Registration in main server switch statement: maps 'extract_assets' tool call to handlers.extractAssets with parsed arguments. handlers imported from './tools/handlers/index.js'.
    case "extract_assets":
      result = await handlers.extractAssets(this.ctx, args.file_key, args.page_name, args.frame_name, args.output_dir || "./figma-assets");
      break;
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of behavioral disclosure. It does an excellent job describing key behaviors: progress tracking with specific output format, automatic categorization into folders, smart naming, composite group handling, and final summary. It even provides implementation details ('Look for "isCompositeAsset: true"'). The only minor gap is lack of explicit mention about permissions or error handling.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-structured with clear sections ('HOW IT WORKS', 'TYPICAL WORKFLOW') and uses bullet points effectively. While slightly longer than minimal, every sentence adds value. The front-loading of the core purpose is excellent, though the workflow section could be slightly more concise.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex asset extraction tool with no annotations and no output schema, the description provides substantial behavioral context. It explains the extraction process, categorization, naming, progress reporting, and provides a workflow. The main gap is the lack of explicit information about return values/output format, though the 'final summary' hint partially addresses this.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents all parameters. The description doesn't add any parameter-specific information beyond what's in the schema (e.g., it doesn't explain format requirements for file_key or matching behavior for page_name/frame_name). However, it does provide context about the output_dir's purpose through the workflow example.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the specific action ('extract all assets') and resource ('from a frame'), distinguishing it from siblings like 'extract_styles' or 'get_screenshot'. It explicitly mentions what types of assets are extracted (icons, images, composites) and how they're processed, providing clear differentiation from other tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The 'TYPICAL WORKFLOW' section explicitly states when to use this tool ('extract_assets → download all') and provides a clear prerequisite ('get_frame_info → see what assets exist and identify composite groups'). It also implies when not to use it (e.g., for styles extraction or screenshots, which are handled by sibling tools).

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/alucardeht/figma-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server