TypeUI
io.github.bergsidePublished by GitHub account bergsidev1.0.1Updated Oct 25, 2026
Design systems, UI prompts, and layout variations for AI coding tools.
The publisher's own server: type.com
Overview
Hosted TypeUI MCP server that gives AI coding tools design systems, UI prompts and layout variation guidance for building interfaces.
- What you can do
- Supply curated design skills and design-system guidance while the assistant writes or refactors UI
- Provide UI prompts and layout variations for common page sections such as hero, pricing, feature, CTA, testimonial and footer blocks
- Offer UI/UX fundamentals guidance on spacing, typography and color contrast
- Point to a companion CLI for generating, updating, randomizing and pulling SKILL.md or DESIGN.md files locally
- When to use it
- Useful when you build websites or app screens with an AI coding tool and want the generated UI to follow a consistent design system or a specific visual style instead of generic defaults.
- Requirements
- A remote MCP endpoint at mcp.typeui.sh over streamable HTTP; OAuth authentication is required. No environment variables or headers are declared. The optional CLI runs through npx and needs Node.js.
Installation
In SourceWeft
- Open TypeUI in the dashboard and add it to a workspace.
- Connect the required credentials. They are stored privately and never shown on public pages.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Use with an agent
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/io.github.bergside%2Ftypeui/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Other MCP clients
Add this to your client's mcpServers config. The client completes OAuth sign-in on first connect.
{
"mcpServers": {
"typeui": {
"type": "http",
"url": "https://mcp.typeui.sh/mcp"
}
}
}README
TypeUI - Build better UI with AI
TypeUI is an AI-first platform where you can use design skills, UI prompts, and other resources to enhance AI capabilities to generate better user interfaces and design when building websites and applications. The resources from this repository will work with any AI provider, given they support markdown files or general prompting.
Getting started
Install TypeUI where you build. Connect your AI coding tool to TypeUI MCP, then ask for UI in plain language while TypeUI supplies design systems, prompts, and layout variation guidance.
Design skills
Check out all design skills. Available in both DESIGN.md and SKILL.md formats.
UI design prompts
Browse all major UI prompt categories from TypeUI Prompts.
UI/UX Fundamentals
Use our official skill for UI/UX fundamentals based on years of experience, knowledge and data from the design world to make sure that any UI you generate together with the design skills and UI prompts from TypeUI will keep fundamental design principles in order, such as spacing, typography, color contrasts, and more.
Run this command to install it locally:
CLI commands
All commands run via npx typeui.sh. Add --help to any command to see its options.
generate
Run interactive prompts and generate a fresh SKILL.md (default) or DESIGN.md file.
update
Run interactive prompts and update existing output files in place.
randomize
Generate a fully randomized local design system — useful for quickly seeding a starter spec.
list
Browse available registry specs from bergside/awesome-design-skills (with typeui.sh preview links), then pull one automatically.
pull <slug>
Pull a registry markdown file by slug. SKILL.md writes to provider paths; DESIGN.md writes to the project root.
Local development
If you want to use this locally these are the commands you need to run:
Then use the following command instead of npx in your terminal:
License
The CLI and public registry is open-source under the MIT License.
Pro version
Get access to enhanced design skill files, UI prompts for layouts and a private Discord community by getting the pro version and thus supporting our open-source work.
Sponsors
A huge thank you to the companies supporting our open-source work.
Want to see your logo here? Become a sponsor.
Source: README.md at commit efc5f21
Tools
0Version history
1- v1.0.1LatestSep 16, 2026
Score
75% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 daysRequired25 pts
- Auth declaredRequired10 pts
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Declare a license in the repository or the registry entry.
Other Type integrations
type.com
com.typeOfficial publisher
Remote MCP server for a shared workspace platform, letting an assistant work with spaces, threads, documents, skills, apps, automations and connected tools.

CellType UDI
Published by GitHub account celltype
Search public biomedical datasets, measured results, papers, trials and drug evidence with sources.
TypeSafe
com.foreverworks
Get calibrated yes or no answers, choices and scores from TypeSafe's Jev, for one item or 1000 rows.
PaperType
dev.papertype
PaperType renders Markdown, HTML or invoice, certificate and report templates to PDF files through the PaperType API, saving them locally.

TypeSafe Jev
Published by GitHub account PerryLink
A local MCP server over TypeSafe Jev that lets an assistant ask, rank and check judgments and receive calibrated probabilities.