TanStack MCP Server
Wraps the TanStack CLI to provide programmatic access to documentation, library listings, and project scaffolding for the TanStack ecosystem. It enables AI assistants to search documentation and create new TanStack applications through the Model Context Protocol.
README
@g7aro/tanstack-mcp
MCP server that wraps the TanStack CLI to provide programmatic access to TanStack documentation, libraries, add-ons, ecosystem partners, and project scaffolding.
Built as a drop-in replacement after TanStack removed the built-in MCP server from @tanstack/cli.
Tools
| Tool | Description |
|---|---|
listTanStackAddOns |
List all available TanStack Start add-ons for a given framework |
getAddOnDetails |
Get detailed info about a specific add-on (files, deps, options, routes) |
createTanStackApplication |
Scaffold a new TanStack Start app with add-ons and options |
tanstack_list_libraries |
List all TanStack libraries with descriptions and links |
tanstack_doc |
Fetch the full content of a TanStack documentation page |
tanstack_search_docs |
Search across TanStack documentation |
tanstack_ecosystem |
List TanStack ecosystem partners with optional filters |
Quick Install
Auto-detect installed AI clients and register the MCP server in one command:
npx @g7aro/tanstack-mcp --install
This will detect and configure all supported clients on your machine.
Options
npx @g7aro/tanstack-mcp --install # Interactive — pick which clients
npx @g7aro/tanstack-mcp --install --all # Install into all detected clients
npx @g7aro/tanstack-mcp --install cursor codex # Install into specific clients only
npx @g7aro/tanstack-mcp --uninstall # Remove from all clients
Supported clients
| Client | Detection | Config method |
|---|---|---|
| Claude Code | claude CLI |
claude mcp add |
| Codex (OpenAI) | codex CLI |
codex mcp add |
| Cursor | ~/.cursor/ dir |
~/.cursor/mcp.json |
| Windsurf | ~/.windsurf/ dir |
~/.windsurf/mcp.json |
| Trae | ~/.trae/ dir |
~/.trae/mcp.json |
| Antigravity | ~/.gemini/antigravity/ dir |
mcp_config.json |
| OpenCode | ~/.config/opencode/ dir |
opencode.json |
| Zed | ~/.config/zed/settings.json |
settings.json |
| VS Code (Copilot) | settings.json / code CLI |
settings.json |
Manual Setup
If you prefer to configure manually, add to your client's MCP config:
{
"mcpServers": {
"tanstack": {
"command": "npx",
"args": ["-y", "@g7aro/tanstack-mcp"]
}
}
}
Tools
| Tool | Description |
|---|---|
listTanStackAddOns |
List all available TanStack Start add-ons for a given framework |
getAddOnDetails |
Get detailed info about a specific add-on (files, deps, options, routes) |
createTanStackApplication |
Scaffold a new TanStack Start app with add-ons and options |
tanstack_list_libraries |
List all TanStack libraries with descriptions and links |
tanstack_doc |
Fetch the full content of a TanStack documentation page |
tanstack_search_docs |
Search across TanStack documentation |
tanstack_ecosystem |
List TanStack ecosystem partners with optional filters |
Prerequisites
- Node.js >= 18
npxavailable in PATH (ships with npm)
How it works
Each MCP tool maps to a @tanstack/cli command with --json output:
listTanStackAddOns -> tanstack create --list-add-ons --framework <f> --json
getAddOnDetails -> tanstack create --addon-details <id> --framework <f> --json
createTanStackApplication -> tanstack create <name> --framework <f> --add-ons <a,b> ...
tanstack_list_libraries -> tanstack libraries --json
tanstack_doc -> tanstack doc <library> <path> --json
tanstack_search_docs -> tanstack search-docs "<query>" --json
tanstack_ecosystem -> tanstack ecosystem --json
The server spawns npx @tanstack/cli for each invocation, parses the JSON output, and returns it through the MCP protocol over stdio.
Development
npm install
npm run build # compile TypeScript -> dist/
npm start # run the server (stdio)
npm run dev # watch mode
License
MIT
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。