- Home
- Skills
- Agents & Orchestration
- figma bridge
Works with the AI Agents you already use
figma bridge
Read live Figma document structure, CSS, and tokens directly from Figma desktop via a local MCP bridge.
Free
figma bridge
Example session with this skill installed
Check my Figma selection and tell me the CSS for this button. Also, export it as an SVG.
- Read your context and instructions
- Compiled the figma bridge
- Generated the design tokens stylesheet
I have connected to the Figma file and located your current selection: a component named "Primary-Button-Submit" (ID 42:89).
Design Data for Primary-Button-Submit
Computed CSS
/* Figma Computed Styles */
display: flex;
padding: 12px 24px;
justify-content: center;
align-items: center;
gap: 8px;
border-radius: 6px;
background: var(--colors-brand-primary, #0052FF);
box-shadow: 0px 2px 4px 0px rgba(0, 0, 0, 0.10);
Typography
- Text: "Submit Request"
- Font: Inter, Semi Bold (600)
- Size: 14px
- Line Height: 20px (Auto)
Exported Assets
- SVG: Saved to
figma-bridge-out/primary-button-submit.svg
Next steps
- Review the generated SVG in the
figma-bridge-out/directory. - I can now generate a Tailwind or CSS module component based on these styles.
- If you need the hover state variables, please select the hover variant in Figma and ask me to inspect it.
figma-bridge-tokens.css
CSS · design tokens
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop guessing design specs from low-resolution screenshots or manual descriptions. This skill connects your agent directly to Figma desktop, allowing it to inspect the live document structure, extract CSS, and read design tokens without needing a paid Figma API plan or access tokens.
What it does
- Inspect nodes to read the exact hierarchy, layout properties, and dimensions of any Figma layer.
- Extract CSS directly from components to get computed styles, including flexbox, padding, and borders.
- Read text and styles to grab every copy string, font family, and weight with 100% accuracy.
- Pull design tokens by accessing variables and styles for consistent implementation of color and spacing systems.
- Export assets to generate PNG or SVG files for icons and components directly into your local project root.
- Visual confirmation by capturing images of specific frames to align code implementation with the visual intent.
How it works
- Connect the agent to Figma by calling
figma_pingto verify the local bridge relay and plugin are active. - Locate components or frames using
figma_findor by reading the user's currentfigma_selection. - Analyze the node depth and properties using
figma_nodeandfigma_cssto gather implementation data. - Implement by converting the retrieved JSON and style data into clean code or exported assets.
Frameworks & tools
This skill utilizes a local relay server and the Figma desktop application. It interfaces via MCP (Model Context Protocol) tools or a CLI wrapper using Node.js/npx.
Why this beats prompting it yourself
Generic agents are blind to the underlying document tree and design variables, leading to "hallucinated" CSS and approximate layouts. This skill provides the actual data structure, ensuring your code matches the source of truth in Figma.
Use cases
- Convert a Figma component into a pixel-perfect React or Tailwind component.
- Scaffolding design systems by exporting variables into CSS or SCSS files.
- Auditing existing code against the latest design file for visual regressions.
- Exporting a batch of icons as optimized SVGs directly into your
src/assetsfolder.
Known limitations
Read-only access. It cannot edit Figma files. Requires Figma desktop and the figma-bridge plugin to be open locally. Output is restricted to the defined project root.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Read-only access only. Requires Figma desktop and the bridge plugin to be running locally. Large data outputs are saved to local files rather than returned in the chat context.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 11 days ago
- Free with an Agensi account
- Software
- Without Tools MCP@ubuligan/figma-bridge
Node.js