What runtime does heroicons-mcp require?
It requires Git and Bun for development, but can be run via `npx` (Node.js) in production.
@SeeYangZhi
基本信息
heroicons-mcp is a Model Context Protocol (MCP) server that exposes Heroicons. It allows LLMs and agentic applications to search, list, and generate usage examples of icons in React applications.
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"heroicons": {
"command": "npx",
"args": [
"heroicons-mcp",
"--stdio"
]
}
}
}Search for icons from heroicons by name or category
Get usage examples for an icon
List all icons from the heroicons library, optionally filtered by style
heroicons-mcp is a Model Context Protocol (MCP) server that exposes Heroicons as resources and tools for LLMs and agentic applications. Built with Bun and the MCP TypeScript SDK, it provides access to Heroicons in Outline and Solid styles.
Install dependencies with bun install, then run either npx heroicons-mcp for HTTP mode (defaults to port 3000) or npx heroicons-mcp --stdio for stdio mode. For integration with Claude Desktop, add the server to the mcpServers section of the Claude Desktop configuration file using the npx command or a direct path to the built entry point.
list_all_icons tool with optional style filtersearch_icons tool by name or keywordsget_icon_usage_examples tool returning JSX codeIt requires Git and Bun for development, but can be run via `npx` (Node.js) in production.
Use `npx heroicons-mcp` for HTTP mode (default), or add `--stdio` for stdio mode.
Add it to your Claude Desktop config file under `mcpServers` using the `npx` command or a direct path to the built `entry.js` file with the `--stdio` flag.
It exposes three tools: `list_all_icons`, `search_icons`, and `get_icon_usage_examples`.
The server includes icons in both Outline and Solid styles from Heroicons.
Painless E2E Automation for Mobile and Web
A search tool helps dev to solve the naming things problem.
AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.
🚀 The fast, Pythonic way to build MCP servers and clients.
评论