aceternity-mcp
MCP server for Aceternity UI components, enabling AI assistants to discover, search, install, and generate page layouts with 106 components.
README
<div align="center">
Aceternity MCP
Model Context Protocol server for Aceternity UI components
Discover, search, and install 106 Aceternity UI components from your AI assistant. Generate full page layouts, get context-aware recommendations, and install everything with a single command.
</div>
Install
pipx install aceternity-mcp
aceternity-mcp install
Restart your AI tool. Done.
What It Does
Your AI assistant gets 11 tools for working with Aceternity UI:
| Tool | What it does |
|---|---|
generate_page_layout |
Generate a full page layout with sections, components, and install commands |
search_components |
Full-text search across names, descriptions, tags, categories |
recommend_components |
Get recommendations for a use case or design goal |
recommend_combination |
Get component suggestions for each page section |
match_components_to_project |
Match components to a whole project description |
list_components |
List all 106 components with optional filters |
list_categories |
Show all 17 categories with component counts |
get_component |
Get full metadata for a single component |
get_category |
Get all components in a category |
install_component |
Get install command, dependencies, and setup steps |
filter_by_scores |
Filter by visual intensity, animation, performance, etc. |
Page Layout Generator
The standout feature. Ask your AI to generate a complete page layout and it returns:
- Ordered sections with recommended components for each role
- Design tone detection (dark, premium, modern, minimal, etc.)
- Section-specific scoring using 17 role profiles
- De-duplicated components across sections
- Implementation notes for every section
- Batch install command for all components at once
- Performance estimate and dependency summary
7 page types: landing, saas, dashboard, portfolio, ecommerce, blog, documentation
Example prompts:
"Generate a page layout for a dark premium AI SaaS landing page"
"Build me a fintech dashboard layout with analytics"
"Create a portfolio page with 3D effects and scroll animations"
"Design an e-commerce store layout for a clothing brand"
Scoring signals per component:
| Signal | Description |
|---|---|
| Section-role category match | 17 role profiles with preferred categories |
| Section-role tag match | Role-specific tag boosting |
| Section score weights | Tuned dimension weights per role |
| Archetype context | Page-level scoring (SaaS, dashboard, portfolio, etc.) |
| Text relevance | User description keyword matching |
| Design tone coherence | Boosts components matching detected tones |
| Dependency reuse | Bonus for shared npm dependencies |
Example Prompts
"Show me subtle dark background effects with low performance impact"
"Recommend a navbar + hero + CTA combination for a SaaS landing page"
"Find highly customizable card components for testimonials"
"Give me low-motion components suitable for a dashboard"
"Install Spotlight and show me what dependencies are needed"
"What components work well for dark mode dashboards?"
"Search for animated text effects"
"Filter components with visual intensity above 7 and low performance impact"
Supported AI Tools
| Tool | Auto-configured |
|---|---|
| Claude Desktop | Yes |
| Claude Code CLI | Yes |
| Cursor | Yes |
| Cline (VS Code) | Yes |
| Windsurf | Yes |
| OpenCode | Yes |
Run aceternity-mcp install and all detected tools are configured automatically.
Manual Configuration
All tools use the same server command:
{
"mcpServers": {
"aceternity-ui": {
"command": "aceternity-mcp-server",
"args": []
}
}
}
| Tool | Config file |
|---|---|
| Cursor | ~/.cursor/mcp.json |
| Claude Code | ~/.claude/mcp.json |
| Cline | VS Code extension settings |
| Windsurf | ~/.codeium/windsurf/mcp_config.json (uses mcp_servers / aceternity_ui) |
| OpenCode | ~/.opencode/mcp.json |
CLI Commands
| Command | Description |
|---|---|
aceternity-mcp install |
Interactive setup wizard |
aceternity-mcp status |
Show installation health |
aceternity-mcp update |
Check for and install updates |
aceternity-mcp repair |
Fix registry, configs, or permissions |
aceternity-mcp diagnose |
Output JSON diagnostics |
aceternity-mcp uninstall |
Remove from all AI tools |
aceternity-mcp --version |
Show version |
# Repair only the registry
aceternity-mcp repair --registry
# Repair only client configs
aceternity-mcp repair --configs
# Non-interactive update
aceternity-mcp update -y
Alternative Install Methods
Docker
docker pull ghcr.io/devinoldenburg/aceternity-mcp
docker run -i ghcr.io/devinoldenburg/aceternity-mcp
npm (wrapper)
npx aceternity-mcp-server
Requires the Python package installed separately (pipx install aceternity-mcp).
pip
pip install aceternity-mcp
From source
git clone https://github.com/devinoldenburg/aceternity-mcp.git
cd aceternity-mcp
pipx install .
Component Registry
106 components across 17 categories:
| Category | Count | Examples |
|---|---|---|
| Backgrounds | 18 | Aurora, Beams, Grid, Spotlight, Vortex |
| Cards | 15 | 3D Card, Card Hover, Expandable Card |
| Text Effects | 13 | Typewriter, Text Generate, Gradient Text |
| Scroll Effects | 11 | Parallax, Scroll Reveal, Tracing Beam |
| Visualization | 9 | Globe, World Map, SVG Mask |
| Hero Sections | 8 | Hero Parallax, Spotlight, Lamp Effect |
| Navigation | 7 | Floating Dock, Navbar Menu |
| Layout | 81 | Bento Grid, Container Scroll, Timeline |
| Forms | 6 | Input, Placeholders, Multi-Step |
| 3D Effects | 5 | 3D Card, 3D Pin, Globe |
Each component includes: detailed description, purpose, behavior, visual characteristics, design tone, difficulty level, dependencies, install command, compatibility flags, and 7 scoring dimensions.
Scoring Dimensions
Every component is rated 1-10 on:
| Dimension | What it measures |
|---|---|
| Visual Intensity | How visually dominant (1 = subtle, 10 = dramatic) |
| Animation Intensity | Amount of motion (1 = static, 10 = heavy animation) |
| Layout Importance | Structural criticality (1 = decorative, 10 = essential) |
| Customization Ease | Flexibility (1 = rigid, 10 = very configurable) |
| Landing Page Fit | Suitability for landing pages |
| Dashboard Fit | Suitability for dashboards |
| Performance Impact | Runtime cost (1 = lightweight, 10 = heavy) |
Development
git clone https://github.com/devinoldenburg/aceternity-mcp.git
cd aceternity-mcp
python -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"
# Run tests
pytest tests/ -v
# Lint + type check
ruff check src/ tests/
mypy src/aceternity_mcp/ tests/
# Validate registry
python scripts/validate_registry.py
Quality
- 172 tests, 9 archetypes, 17 section role profiles, 7 page templates
- ruff lint + format, mypy strict, bandit security scan
- CI: quality gates on every push, auto-publish on release
Security
- Metadata only: No Aceternity source code, documentation, or images in this repository
- No secrets: No API keys or credentials
- Isolated: pipx provides sandboxed virtual environments
- Install commands: Point to Aceternity's official
@aceternityshadcn registry
Troubleshooting
Command not found -- Run pipx ensurepath and restart your terminal.
Registry not found -- Run aceternity-mcp repair or pipx reinstall aceternity-mcp.
MCP server not connecting -- Run aceternity-mcp diagnose and check the output. Repair configs with aceternity-mcp repair --configs.
Update issues -- Run pipx upgrade aceternity-mcp or pipx reinstall aceternity-mcp.
License
MIT -- see LICENSE.
Links
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。