The professional design tool — and an MCP server that lets an AI design in it without you opening it.
Pick this whenReal design work, or when you want an agent to produce design artifacts programmatically.
What makes it different
Its MCP server was BC5's headline possibility demo — design without opening Figma
The industry standard, so its file format is the lingua franca of handoff
What it does
Designs interfaces collaboratively
Exposes an MCP server so assistants can create and read designs
What it does not do
Common wrong expectationsDoesn't suit quick throwaway visuals the way Canva or Excalidraw do
Where it sits in a stack
Professional design, and an example of how far MCP reaches into specialist tools.
ShelfDesign & whiteboarding
Our status · not-usingNo design tooling in the pipeline.
For your projectsThe Figma MCP is the best available argument for why MCP matters — worth citing when explaining the concept to anyone.
| Session | Coverage | Moment | What happened |
|---|
| Session 2: How to Plan, Scope and Sell your AI Projects | demonstrated | 1:16:57 | His team's system-design tool — the lead-qualification engine diagram shown was built in it; Canva/Excalidraw/Miro named as peers |
| Day 2: The Image/Video Model Landscape, Cinematic Prompting, Cost Strategy, and a Commercial Built Live | demonstrated | | Storyboard template; lock the layer |
| Grounding an AI-Built App: Anthropic API Key, Context7 with shadcn, a Figma MCP-Ready Design System, and Deploying to Vercel | demonstrated | | |
| Grounding an AI-Built App: Anthropic API Key, Context7 with shadcn, a Figma MCP-Ready Design System, and Deploying to Vercel | demonstrated | | |
| Pre-Launch Design Testing with an AI Review Agent: Audits, Copy Review, A/B Comparison, Fix Loop into Claude Code, and Ad/Video Scoring | demonstrated | | |
| Design and Figma Overview: the Target Screens and How to Read the Figma File | demonstrated | | |
| App Setup: Canvas and Grid, Settings Every App Needs, Typography and Colour Variables, Responsive Text | demonstrated | | |
| Building the Homepage, Part 2: Two-Column Feature Sections, Text and Button Styles, Image Compression, Checklist Rows, and a Pre-built Footer | demonstrated | | |
| Chapter 3: The Workflow — Four Steps, Four Agreement Rounds, No Going Back | demonstrated | | Local variables carry the whole Visual Direction phase: palette option sets, color modes, font pairs |
| Chapter 4: The Craft — Color, Type, Spacing, and the Tricks You Earn Afterward | demonstrated | | Live surgery throughout: spacing fixes, image temperature/tint/blending modes, variables for spacing tokens |
| Chapter 5a: The Client Build — Discovery to Direction with Lucy's Bakery | demonstrated | | Free-tier build: Auto Layout sitemap boxes, Pages, duplication discipline, Prototype connectors, offset normalization |
| Chapter 5b: The System Build — Styles, Variables, Components, and the Stylekit Pipeline | demonstrated | | The deep-feature tour: Styles, Number Variables, Components, Variants, Sections, Constraints, Wrap, Fill/Hug |
| Chapter 5c: The Finish — Details, Feedback, AI Images, and the Polished Preview | demonstrated | | The finishing toolkit: Open Overlay, wrap+fill, Outline Stroke, Pathfinder Union, Copy/Paste Properties, calculation fields, constraints, spellcheck |
| Section 1: Orientation — What Elementor Really Is (and What It Sits On) | demonstrated | | Free tier; course design distributed as a copyable file |
| Section 3: The Container — Layout, Wrapping Math, and Custom Positioning | demonstrated | | 12-column grid as the design-side source of the 1120 doctrine and the percent sheet |
| Section 4: The Design System — Figma Export, Clean WordPress, and the Class System | demonstrated | | Export workflows: hex harvesting, vector/pixel classification, bounding-box shadow control, 2x exports |
| Basecamp 5: MCP | explained | 0:47:49 | Possibility example: design/wireframe from Claude without opening Figma |
| Chapter 1: The Mindset — Figma Is a Sandbox, and Your Brain Is the Obstacle | explained | | The design sandbox — two lessons of boundary-setting before any usage |
| AI for Designers: Product Research - Personas, Jobs to Be Done, Competitors and a Locked Value Proposition in Claude and ChatGPT | mentioned | | |
| Chapter 2: Why We Struggle — Style Follows Function, and Assumptions Are the Enemy | mentioned | | Named with Adobe XD and Photoshop as tools where an extra page costs nothing - the free-experimentation discipline argument |
| Vibe Coding for Non-Coders: Pain Point to MoSCoW to Stack, 'Use AI to Build AI' with Grok, and an Audio Note-Taker Built Live on Lovable + Supabase + OpenAI | mentioned | | Design import into Lovable |
Figma