M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
T

TypeUI

Community
Build better UI with AI
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #295 of 438
Stars
★ 2.1k Very popular
Transport
Streamable HTTP
Runtime
Hosted (no local runtime) · Node.js
Credentials
No credential needed
License
Other / unspecified
Last commit
46FMRS · D

TypeUI provides design systems, UI prompts, and layout variations for AI coding tools through a Streamable HTTP MCP server. The source does not enumerate tools, authentication, or a client configuration example. The repository metadata lists NOASSERTION for the license, while the README states that the CLI and public registry are open source under the MIT License.

Strongest · Setup experience 12/20 Weakest · Reliability 7/20

Reliability
7/20
Security and permissions
8/20
Maintenance
8/20
Documentation
11/20
Setup experience
12/20
Why each score
Reliability 7/20
The manifest provides only one streamable-HTTP remote endpoint. The source contains no tool list, initialization or tool-handshake evidence, tests, CI, or error-handling evidence, so a normal path is merely plausible; reproducibility, edge-case coverage, and diagnosability are unverified and deducted.
Security and permissions 8/20
There is no evidence of malware, credential theft, or dangerous default side effects. However, the remote service's data flows, authentication, permission scope, third-party network access, and retention practices are not disclosed; least-privilege boundaries and confirmation mechanisms are unverified and deducted.
Maintenance 8/20
The repository is not archived, has 2 open issues, and the manifest includes a version. The README states that the CLI and public registry are MIT-licensed; however, the supplied repository metadata says NOASSERTION, and there is no evidence of commit history, release cadence, dependency updates, a security-response channel, or maintainer commitments, so points are deducted.
Documentation 11/20
The README gives substantial coverage of the product purpose, many client integrations, design skills, prompt categories, and CLI commands. It lacks an MCP tool/parameter inventory, authentication details, service limits, cost information, concrete configuration examples, and troubleshooting; the remote capabilities also cannot be fully verified from the supplied documentation, so points are deducted.
Setup experience 12/20
The manifest clearly supplies an HTTPS streamable-HTTP endpoint, and the README lists many client guides plus CLI installation commands. But concrete client configurations, runtime prerequisites, authentication steps, and connection-validation evidence are mainly delegated to external links; the supplied source does not provide a fully reviewable end-to-end setup path, so points are deducted.

Static review · not runListed 2026-08-14

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • Developers building websites or applications with AI coding tools
  • Teams seeking design systems, interface prompts, and layout references
  • AI users whose providers support markdown files or general prompting workflows

Not for

  • Users who require a documented list of MCP tool names and parameters
  • Users seeking traditional design software without an AI coding workflow
  • Scenarios requiring authentication, permissions, or private-resource guarantees not documented in the source

Required permissions

  • The client needs access to the remote MCP endpoint at https://mcp.typeui.sh/mcp
  • Local CLI workflows require installing dependencies and writing generated SKILL.md or DESIGN.md files
  • The source does not specify authentication requirements for the remote server

Risks and side effects

  • The server manifest does not enumerate MCP tools, so its callable capabilities cannot be confirmed from the source
  • No authentication or data-handling details are provided
  • Generated design guidance still requires human review against project requirements and UI/UX fundamentals
  • The Pro version includes enhanced design skills, layout prompts, and a private Discord community, so some resources may not be public

Setup

Before you start

Runtime：Hosted (no local runtime) · Node.js

Connect an MCP-capable AI coding tool to https://mcp.typeui.sh/mcp and follow the TypeUI guide for that client. For local UI/UX fundamentals, run npx skills add https://github.com/bergside/typeui --skill typeui-fundamentals.

Check that it works

After connecting your client to the hosted endpoint https://mcp.typeui.sh/mcp, issue a plain-language request (e.g. "give me a Bento-style hero layout") and confirm TypeUI returns design system, prompt, or layout guidance; you can also run npx typeui.sh generate to confirm the CLI produces a SKILL.md/DESIGN.md file.

Troubleshooting

  1. Confirm that the client supports Streamable HTTP MCP connections
  2. Check that the client uses https://mcp.typeui.sh/mcp as the server URL
  3. Follow the TypeUI guide for the selected AI tool in the README
  4. For local CLI use, run npm install and npm run build, then use node dist/cli.js --help
  5. Verify that the selected file format and target path match the SKILL.md or DESIGN.md workflow

Things to try

Once connected, you can ask your AI assistant things like:

  • Generate a login page UI using the Glassmorphism design skill
  • Show me several layout variations for a hero section
  • Refactor my pricing page using the Bento design system
  • Check my page's spacing and typography against TypeUI UI/UX fundamentals

Use cases

Generate design direction for websites and applications
Access design systems and UI prompts
Explore layout variations for marketing and application pages
Add UI/UX fundamentals such as spacing, typography, and color contrast to AI-generated interfaces
Use the TypeUI CLI to generate, update, randomize, or pull local design files

Supported clients

CodexPartial support
ClaudePartial support
CursorPartial support
GrokPartial support
Z.aiPartial support
HermesPartial support
OpenClawPartial support
EvePartial support
MistralPartial support
OpenCodePartial support
VS CodePartial support
AntigravityPartial support
WindsurfPartial support
JetBrains IDEsPartial support
ZedPartial support
GitHub CopilotPartial support
ClinePartial support
QwenPartial support
AmpPartial support
GoosePartial support

Listed from the project's documentation, not tested by this site.

Overview

TypeUI is an AI-first platform that helps AI coding tools generate better website and application interfaces through design skills, UI prompts, and layout-variation guidance. Its resources work with AI providers that support markdown files or general prompting. The MCP server is available through a remote Streamable HTTP endpoint.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 63.3k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 2a977f1f6616 Data synced 2026-10-19 Read the FMRS scoring method