Lucide Icons MCP
lucide-icons-mcp
README
lucide-icons-mcp
A Model Context Protocol (MCP) server exposing Lucide React icons as resources and tools for LLMs and agentic applications. Built with Bun and the MCP TypeScript SDK.
What is Lucide?
Lucide is a beautiful & consistent icon toolkit made by the community. It's an open-source icon library that provides over 1,500+ carefully crafted icons in a single, consistent style. Originally forked from Feather Icons, Lucide has grown into one of the most popular icon libraries for modern web development.
What is MCP?
The Model Context Protocol (MCP) is a standard for AI tools to request specific context from sources outside their main training data.
This MCP server allows AI coding assistants and other agentic applications to access information about Lucide React icons, enabling better assistance with icon search, discovery, and implementation.
Features
- 🔍 Icon Search: Search through 1,500+ Lucide icons by name or category
- 📂 Category Browsing: List icons by categories (Design, Communication, Media, etc.)
- 💡 Usage Examples: Get React/JSX code examples for any icon
- 🔧 Icon Information: Detailed information about each icon
- 🚀 MCP Integration: Ready for Claude Desktop and other MCP clients
- 🌐 Dual Mode: HTTP server or stdio-based MCP server
- 📊 Comprehensive Coverage: All Lucide icons with proper JSX usage
Prerequisites
Getting Started (Development)
1. Clone the repository
git clone https://github.com/SeeYangZhi/lucide-icons-mcp.git
cd lucide-icons-mcp
2. Install Bun (if you don't have it)
Refer to the official Bun installation guide.
After installation, restart your terminal and check:
bun --version
3. Install dependencies
bun install
4. Build the project
This compiles the TypeScript source to JavaScript in the build directory.
bun run build
Usage
HTTP Mode
You can run the HTTP server using npx:
npx lucide-icons-mcp
This starts the HTTP server (defaults to port 3000).
Or install globally:
npm install -g lucide-icons-mcp
Then run:
lucide-icons-mcp
Stdio Mode
npx lucide-icons-mcp --stdio
# or if installed globally
lucide-icons-mcp --stdio
Local Development
There are two main ways to run the MCP server:
1. HTTP Mode
Suitable for clients that support communication over HTTP.
For development (using Bun):
bun run start
# or directly
bun run src/entry.ts
2. Stdio Mode
Often used for direct integration with tools like Claude Desktop or the MCP Inspector.
For development (using Bun):
bun run src/entry.ts --stdio
Configuration with AI Tools
Example: Claude Desktop
To use this MCP server in Claude Desktop:
- Open your Claude Desktop configuration file:
code ~/Library/Application\ Support/Claude/claude_desktop_config.json
- Add the server to the
mcpServerssection:
Option A: via npx (Recommended):
{
"mcpServers": {
"lucide-icons": {
"command": "npx",
"args": ["lucide-icons-mcp", "--stdio"]
}
}
}
Option B: Pointing directly to the build output:
{
"mcpServers": {
"lucide-icons": {
"command": "node",
"args": ["/ABSOLUTE/PATH/TO/lucide-icons-mcp/build/entry.js", "--stdio"]
}
}
}
- Save the file and restart Claude Desktop.
- You should now see the "lucide-icons" server available in Claude's tools panel.
Tools Available (MCP)
This MCP server exposes the following tools to AI coding assistants:
1. search_icons
- Description: Search for Lucide icons by name or category
- Parameters:
query(string): Search term for icon name or categorycategory(optional string): Filter by specific categorylimit(optional number): Maximum results to return (default: 20)
2. search_categories
- Description: Search for icon categories by name
- Parameters:
query(string): Search term for category namelimit(optional number): Maximum results to return (default: 10)
3. get_icon_usage_examples
- Description: Get React/JSX usage examples for a specific Lucide icon
- Parameters:
name(string): Icon name (e.g., 'Home', 'User', 'Settings')
4. list_icons_by_category
- Description: List all icons in a specific category
- Parameters:
category(string): Category name to list icons forlimit(optional number): Maximum results to return
5. list_all_categories
- Description: List all available icon categories with their icon counts
- Parameters: None
6. get_icon_info
- Description: Get detailed information about a specific icon
- Parameters:
name(string): Icon name to get information for
7. list_all_icons
- Description: List all available Lucide icons
- Parameters:
category(optional string): Filter by categorylimit(optional number): Maximum results to return (default: 100)
Example Usage
Here's how an AI tool might use this MCP server:
Example 1: Finding Icons
User: "Find me icons related to ArrowRight"
AI tool calls search_icons:
{
"query": "ArrowRight",
"limit": 5
}
Response: Lists icons like ArrowRight
Example 2: Getting Usage Examples
User: "Show me how to use the ArrowRight icon"
AI tool calls get_icon_usage_examples:
{
"name": "ArrowRight"
}
Response:
import { ArrowRight } from "lucide-react";
function Example() {
return (
<div>
<ArrowRight />
</div>
);
}
Example 3: Browsing Categories
User: "What design-related icons are available?"
AI tool calls search_categories:
{
"query": "design"
}
Then calls list_icons_by_category":
{
"category": "Design",
"limit": 10
}
Icon Categories
Lucide icons are organized into categories such as:
- Accessibility (29)
- Accounts & access (131)
- Animals (22)
- Arrows (204)
- Brands (21)
- Buildings (24)
- Charts (31)
- Communication (50)
- Connectivity (85)
- Cursors (33)
- Design (141)
- Coding & development (233)
- Devices (156)
- Emoji (20)
- File icons (164)
- Finance (51)
- Food & beverage (67)
- Gaming (137)
- Home (49)
- Layout (135)
- Mail (26)
- Mathematics (71)
- Medical (41)
- Multimedia (130)
- Nature (19)
- Navigation (135)
- Notification (39)
- People (3)
- Photography (74)
- Science (30)
- Seasons (4)
- Security (55)
- Shapes (43)
- Shopping (25)
- Social (114)
- Sports (9)
- Sustainability (20)
- Text formatting (240)
- Time & calendar (55)
- Tools (62)
- Transportation (56)
- Travel (63)
- Weather (42)
Testing MCP Locally with Inspector
You can test the MCP server locally using the MCP Inspector.
First, ensure the project is built:
bun run build
Then launch the Inspector:
npx @modelcontextprotocol/inspector node ./build/entry.js --stdio
This opens the Inspector interface for interactive testing of your MCP server.
Development Scripts
bun run dev: Starts the server in HTTP mode for developmentbun run dev:stdio: Starts the stdio MCP server for developmentbun run build: Compiles TypeScript to JavaScript (output inbuild/)bun run lint: Lints the codebase using ESLintbun run lint:fix: Automatically fixes linting issuesbun run crawl: Crawls Lucide website to update icon databun run pre-build: Crawls data, builds icon metadata, and fixes linting
Project Structure
lucide-icons-mcp/
├── src/
│ ├── entry.ts # Main entry point
│ ├── http.ts # HTTP server implementation
│ ├── stdio.ts # Stdio server implementation
│ ├── utils.ts # MCP server and tools logic
│ ├── data/
│ │ └── icons.ts # Generated icon metadata and helpers
│ └── scripts/
│ ├── main.ts # Web crawler for Lucide icons
│ ├── build-data.ts # Icon data processing
│ └── routes.ts # Crawler route definitions
├── build/ # Compiled JavaScript output
├── storage/ # Crawler data storage
└── data/ # Processed icon metadata
Data Sources
This project includes a web crawler that automatically extracts icon information from the Lucide website, ensuring up-to-date icon data and categories.
Resources
- Lucide - The icon library
- Lucide React - React implementation
- Model Context Protocol - MCP specification
- MCP TypeScript SDK - SDK used
- MCP Inspector - Testing tool
- Bun - JavaScript runtime
License
This project is licensed under the MIT License. See LICENSE for details.
The project includes icons from Lucide, which are licensed under the ISC License. See the LICENSE file for full attribution details.
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。