Analyze a Figma file via the Figma REST API and generate an interactive design knowledge graph (pages, screens, components, component sets, instances, design tokens) with a kind:"design" dashboard.
Tasks
Parse the provided Figma URL or file key and optional --language flag from arguments.
Run the bundled figma-scan.mjs script to fetch Figma data and generate a scan-manifest.json.
Dispatch design-analyzer subagents to analyze batches of nodes concurrently.
Run the figma-merge.mjs script to combine intermediate files into a knowledge graph and launch the dashboard.
Inputs
A Figma file URL or bare file key
Optional --language <lang> flag
FIGMA_TOKEN environment variable
Outputs
$UA_DIR/intermediate/scan-manifest.json
analysis-batch-<N>.json files in the intermediate directory
$UA_DIR/knowledge-graph.json
$UA_DIR/meta.json
Limitations and checks
This skill makes outbound calls to api.figma.com and is not fully offline.
The FIGMA_TOKEN must never be written to the graph, meta.json, logs, or intermediate files.
Subagent batch analysis runs up to 5 batches concurrently; failed batches are logged and skipped rather than blocking the process.
Verify that the FIGMA_TOKEN environment variable is set before proceeding; if missing, stop and instruct the user to create and export a token.
Confirm that api.figma.com outbound network calls are expected and inform the user that this skill is not fully offline.
Ensure FIGMA_TOKEN is only transmitted in the X-Figma-Token request header and does not appear in generated files or logs.
Check if packages/core/dist/figma/index.js exists; if missing, install dependencies and build the core package.
If the scan prints UP_TO_DATE, report that the design graph is already up to date and stop execution.
AI Search
Find projects, verify facts, compare options, or turn a complex need into an actionable plan
Try a searchA click only fills the search box; you stay in control