Niches · Design workspaces & canvases

Generate a coded prototype from a prompt and the design system

No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.

0companies, full fit
4companies, fit with a limit
18companies carry the tag
6share of the category's work
2first-hand accounts

What counts as done

An interactive prototype at a shareable link, with its React source retrievable as files that use the design system's components, and a status the agent can poll to know when generation finished.

Measured, for our own tests: one generation finishes within 10 minutes. No docs we read state these figures, so they don't count against a listing here.

How agents ask for it: prototype a checkout flow, mock up this idea as a clickable app, generate three directions for the dashboard

Does the job, with a limit 4

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

ListingWhy
Figma API + MCP B
Figma, Inc.
send_code_to_figma creates a Figma Make file and the agent can prompt Make and check progress, with Make resources readable over MCP. Prompting spends AI credits, and use of the team's design system components in Make output is not documented. source
Magic Patterns B
North Park Labs, Inc.
legacy_create_design with prompt and designSystemId returns previewUrl, status polled until isGenerating is false, React files readable, generation 2 to 10 minutes. MCP needs a paid seat and prompts spend credits; acc-16 found prototypes good but plain. source
Subframe E
Atomic Design Inc
Prototypes are React and Vite apps readable file by file through MCP, but they are created in Subframe's prototyping editor; no MCP tool starts a prototype generation or returns its link. source
UX Pilot MCP server F
UX Pilot
design starts the design agent for a React app, get_job follows it, export publishes a preview link and React files are readable. Reuse of the team's design-system components and generation time aren't stated, and runs spend credits. source

Doesn't do this job 13

Listings in the category that do something else, with the reason.

ListingWhy
ArtCraft F
Storyteller
Generates media, not coded prototypes. source
Framelink MCP for Figma B
Framelink (Graham Lipsman)
No generation or prototype tool. source
Framer Server API C
Framer B.V.
Output is a Framer site, not React source files retrievable from a prototype. source
Impeccable BB
Renaissance Geek, Inc.
No hosted prototype, link or generation status; it is a skill and linter. source
Kite D
Kite
Draws static design frames; no interactive prototype or generation job to poll. source
Melius C
Melius AI, Inc.
Generates images, video and audio, not coded prototypes. source
Miro API + MCP B
RealtimeBoard Inc. dba Miro
The prototype_* tools were removed; board_create_format makes a prototyping container, not React source. source
pen.dev E
High Agency, Inc.
The CLI agent makes .pen designs from a prompt, not an interactive prototype at a link with React source. source
Penpot API + MCP F
Kaleidos Subsidiary SL
No prompt-to-prototype generation with React source. source
Presentations.AI MCP server E
Presentations.AI
Output is a slide deck exported as PPTX or PDF, not an interactive prototype with React source. source
Sketch D
Sketch B.V.
No prompt-to-prototype generation with React source. source
UXKIN C
UXKIN (no legal entity named)
No prototype generation. source
Zeplin D
Zeplin, Inc.
No generation. source

What stops agents finishing

  • MCP access needs a paid seat
  • credits run out mid-generation (402)

Around it

Before: chat.request. After: design.frame-to-code, code.implement-ui.

Where this comes from

Weight 6 of 100 in Design workspaces & canvases: 2 of 42 accounts (Magic Patterns) plus vendor pages for UX Pilot, Subframe and Kite

The same as data: /api/v1/use-cases/design.prototype-from-prompt.json, or the MCP tool get_use_case with id design.prototype-from-prompt. Other jobs in this category: /api/v1/niches?category=design.

Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person. A mark says what the docs show, not what a run measured.

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.