Design and diagrams decision
AI Designer
A capable engineer can reproduce the core prompt→HTML design workflow in about a week, but the full product (MCP integrations, publishing, curated ultradesign quality and production polish) relies on vendor-hosted services and curated workflows that are not practical to replicate in scope.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need - the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship. All AI Designer alternatives, with the arithmetic →
What a replacement has to do
- Prompt → model generates HTML/CSS + images → render editable canvas → save/export design
What it still won’t have
- MCP server integrations and turnkey agent pairing flows
- Pro features like publish/unlimited websites and domain connections
- Unlimited WebGL effects, advanced image tooling, and priority support
- Any proprietary prompt engineering and curated 'ultradesign' workflows
What remains hard
- Product polish and ongoing maintenance
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 5 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a one-user AI design generator web app using Next.js (React), a simple Postgres DB, S3-compatible storage, and OpenAI (or other) REST APIs for text-to-HTML and image generation. Core features in scope: (1) prompt editor with templates and history, (2) server-side API that sends prompts to the model and receives HTML/CSS+image URLs, (3) iframe-based renderer that safely displays generated HTML and exposes a 'refine' textbox to resubmit edits, (4) CRUD for projects and saved brand kits, (5) export endpoint to download HTML/CSS and assets as a ZIP, (6) simple OAuth/email auth, logging, and usage metering. Out of scope: multi-user billing UI, WebGL effects library, advanced MCP agent pairing, and production-grade autoscaling. Provide input validation, HTML sanitization, retries/backoff for model calls, unit tests for API routes, and end-to-end tests for the prompt→render→export workflow.
How we checked
How the score was reached
- Partly verdict base52
- An open-source build was found+5
- 5 cited sources+3
- Price verified on pricing page+3
- 3/3 assessment runs agreed+4
- Evidence score67
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 5
Every page the run actually retrieved.
- official productAIDesigner home
- official pricingAIDesigner pricing
- official docsAIDesigner MCP docs
- open sourceonlook-dev/onlook
- open sourceZSeven-W/openpencil
Integrity checks
What held up, and what did not.
