The fullstack MCP framework to develop MCP Apps for ChatGPT / Claude & MCP Servers for AI Agents.
mcp-use is a full-stack framework for building and deploying Model Context Protocol servers and interactive MCP Apps, available as both a TypeScript npm package and a Python PyPI package. Developers use the SDK to define MCP tools with Zod schemas, then optionally attach React widget components that render inside Claude, ChatGPT, and other MCP clients without rewriting code per platform. A built-in MCP Inspector, available both online and as an open-source package, lets developers test and debug servers locally at a `/inspector` endpoint during development. The `npx create-mcp-use-app` scaffolding command generates a ready-to-run project, and widgets placed in a `resources/` directory are auto-discovered without manual registration. Production deployment is handled through Manufact MCP Cloud, which adds observability, metrics, logs, and branch deployments connected directly from a GitHub repository. Claude Code users can install a dedicated skill via skills.sh to accelerate MCP App development. The framework suits developers building tools that surface structured, interactive UI inside AI chat interfaces.
- ✓Open-source license (MIT)
- ✓Actively maintained (<30d)
- ✓Healthy fork ratio
- ✓Clear description
- ✓Topics declared
- ✓Mature repo (>1y old)
claude mcp add mcp-use -- npx -y create-mcp-use-app{
"mcpServers": {
"mcp-use": {
"command": "npx",
"args": ["-y", "create-mcp-use-app"]
}
}
}MCP Servers overview
<div align="center">
<a href="https://mcp-use.com">
<img alt="mcp-use" src="https://raw.githubusercontent.com/mcp-use/mcp-use/main/docs/logo/banner-mcp-use.webp" width="100%">
</a>
<br /><br />
<div id="user-content-toc">
<ul align="center" style="list-style: none;">
<summary>
<h1>The TypeScript framework for MCP</h1>
<h3>Build, test, and ship MCP servers, ChatGPT plugins, Claude connectors</h3>
</summary>
</ul>
</div>
<p>
Fully Typed, native Views and MCP Apps support, built-in Inspector and first class Agent experience.
</p>
<p>
<a href="https://docs.mcp-use.com/v2/typescript/getting-started/welcome"><strong>Documentation</strong></a>
· <a href="https://inspector.mcp-use.com/inspector"><strong>Inspector</strong></a>
· <a href="#examples"><strong>Examples</strong></a>
· <a href="https://manufact.com"><strong>Deploy</strong></a>
</p>
<p>
<a href="https://www.npmjs.com/package/mcp-use">
<img src="https://img.shields.io/npm/v/mcp-use.svg?label=npm&color=orange" alt="npm version">
</a>
<a href="https://www.npmjs.com/package/mcp-use">
<img src="https://img.shields.io/npm/dw/mcp-use.svg" alt="npm downloads">
</a>
<a href="https://manufact.com">
<img src="https://img.shields.io/badge/made%20by-manufact.com-blue" alt="made by manufact.com">
</a>
<a href="https://github.com/mcp-use/mcp-use/blob/main/LICENSE">
<img src="https://img.shields.io/github/license/mcp-use/mcp-use" alt="MIT license">
</a>
<a href="https://discord.gg/XkNkSkMz3V">
<img src="https://dcbadge.limes.pink/api/server/XkNkSkMz3V?style=flat" alt="Discord">
</a>
</p>
<br /><br />
</div>
> [!NOTE]
> **Migrating from v1? Give it to your agent:**
>
> ```text
> Migrate this mcp-use project to v2 following
> https://docs.mcp-use.com/v2/typescript/server/migration
> ```
>
> [Read the migration guide →](https://docs.mcp-use.com/v2/typescript/server/migration)
## Get started
### Start with your agent
```text
Build an MCP server: https://mcp-use.com/prompt.md
```
[Read the prompt →](https://mcp-use.com/prompt.md)
### Start with code
```bash
npx -y create-mcp-use-app@latest
```
Run `npm run dev` in the generated project · open [`http://localhost:3000/mcp/inspector`](http://localhost:3000/mcp/inspector)
[TS Docs](https://docs.mcp-use.com/v2/typescript/getting-started/welcome)
## Everything you need to ship MCP
<table>
<tr>
<td width="50%" valign="top">
<h3>Fully typed</h3>
<p>Zod schemas flow from tools to structured results, View props, and tool calls.</p>
</td>
<td width="50%" valign="top">
<h3>Native Views</h3>
<p>Bind React Views directly to tools and ship interactive apps without custom extension wiring.</p>
</td>
</tr>
<tr>
<td width="50%" valign="top">
<h3>Agent-first and headless</h3>
<p>Scaffold, invoke, inspect, screenshot, and deploy through your agent.</p>
</td>
<td width="50%" valign="top">
<h3>Built-in debugging tools</h3>
<p>Inspect tools and Views in the browser or headlessly through the CLI.</p>
</td>
</tr>
</table>
## Quickstart
The scaffold gives you the server, TypeScript configuration, development scripts, Inspector, and a React view pipeline. Start it once and the MCP endpoint also serves a client-ready landing page with its connection URL and setup instructions.
Replace its `index.ts` with a view-bound tool like this:
<table><tr><td>
<details>
<summary><strong><code>index.ts</code></strong> — Server entry file for tool definition and metadata</summary>
```typescript
import { MCPServer } from "mcp-use";
import { z } from "zod";
const server = new MCPServer({
name: "weather-app",
title: "Weather App",
version: "1.0.0",
});
const weatherInput = z.object({
city: z.string().describe("City to look up"),
});
const weatherOutput = z.object({
city: z.string(),
temperature: z.number(),
conditions: z.string(),
});
export const getWeather = server.tool(
{
name: "get-weather",
title: "Get weather",
description: "Get the current weather for a city",
inputSchema: weatherInput,
outputSchema: weatherOutput,
view: { name: "weather-card" },
annotations: {
readOnlyHint: true,
destructiveHint: false,
openWorldHint: true,
},
},
async ({ city }) => {
const weather = {
city,
temperature: 22,
conditions: "Sunny",
};
return {
content: [
{
type: "text",
text: `Weather in ${city}: ${weather.conditions}, ${weather.temperature}°C`,
},
],
structuredContent: weather,
};
},
);
export default server;
```
</details>
</td></tr></table>
[Explore MCP server tools →](https://mcp-use.com/docs/typescript/server/tools)
## Add Views to your tools
Create `views/weather-card/view.tsx`. The directory name matches `view.name` on the tool:
<table><tr><td>
<details>
<summary><strong><code>view.tsx</code></strong> — Return a view from your tools: React weather card</summary>
```tsx
import { useCallTool, useToolContext } from "mcp-use/react";
export default function WeatherCard() {
const { status, toolOutput, toolInput } =
useToolContext<"get-weather">();
const refresh = useCallTool("get-weather");
if (status === "pending") {
return <p>Checking the weather in {toolInput?.city ?? "your city"}…</p>;
}
if (status === "error") return <p>Could not load the weather.</p>;
const weather = refresh.data?.structuredContent ?? toolOutput;
return (
<main style={{ padding: 24 }}>
<h2>{weather.city}</h2>
<p>
{weather.temperature}°C · {weather.conditions}
</p>
<button
disabled={refresh.isPending}
onClick={() => void refresh.callTool({ city: weather.city })}
>
{refresh.isPending ? "Refreshing…" : "Refresh"}
</button>
{refresh.error && <p>{refresh.error.message}</p>}
</main>
);
}
```
</details>
</td></tr></table>
<p align="center">
<img src="https://raw.githubusercontent.com/mcp-use/mcp-use/main/static/readme/chatgpt-hello-world.jpg" alt="Hello World MCP App rendered in a ChatGPT conversation" width="100%" />
<br />
<sub>Build interactive UI experiences within ChatGPT with mcp-use.</sub>
</p>
[Build your first MCP App →](https://mcp-use.com/docs/typescript/mcp-apps/quickstart)
## Build
Create the production build:
```bash
npm run build
```
## Inspect
Start development mode to serve the MCP endpoint at [`http://localhost:3000/mcp`](http://localhost:3000/mcp). The Inspector is automatically available at [`http://localhost:3000/mcp/inspector`](http://localhost:3000/mcp/inspector):
```bash
npm run dev
```
<p align="center">
<img src="https://raw.githubusercontent.com/mcp-use/mcp-use/main/static/readme/inspector-hello-world.jpg" alt="Hello World MCP App rendered in the mcp-use Inspector" width="100%" />
<br />
<sub>Invoke tools, validate inputs, and inspect interactive Views in the same development loop.</sub>
</p>
Start a tunnel from the Inspector UI or run `mcp-use dev --tunnel` to get a public URL for your local MCP server and test it with ChatGPT and Claude before deployment. [Learn more about tunneling →](https://docs.mcp-use.com/tunneling)
Inspect the same server headlessly from the terminal, invoke representative tools, and capture a View screenshot:
```bash
npm install --save-dev @mcp-use/client
npx mcp-use client connect local http://localhost:3000/mcp
npx mcp-use client local tools list
npx mcp-use client local tools call get-weather city=Tokyo
npx mcp-use screenshot \
--server local \
--tool get-weather \
city=Tokyo \
--output weather-card.png
```
## Deploy
Ship to [Manufact](https://manufact.com) and get observability, analytics, evals, submission readiness, and Git-based preview environments for free.
```bash
npm run deploy
```
Prefer to run it yourself? Follow the [self-hosting guide →](https://docs.mcp-use.com/typescript/server/deployment/runtime-patterns).
## How mcp-use compares
mcp-use builds on the official TypeScript SDK v2 and adds first-class Views, typed tool-to-UI contracts, an optimized stateless runtime, the Inspector, screenshot verification, agent-first CLI workflows, and deployment.
```mermaid
block-beta
columns 7
metric["Metric"] mcp["mcp-use v2"] fastmcp["FastMCP TS"] official["Official SDK v2*"] xmcp["xmcp"] skybridge["Skybridge"] handler["mcp-handler"]
speed["Speed"] speedMcp["10,982 ops/s"] speedFast["6,628 ops/s"] speedOfficial["8,050 ops/s"] speedXmcp["6,585 ops/s"] speedSkybridge["8,116 ops/s"] speedHandler["6,324 ops/s"]
install["MCP App<br/>dev stack"] installMcp["74.4 MiB"] installFast["122.5 MiB"] installOfficial["99.0 MiB"] installXmcp["121.9 MiB"] installSkybridge["137.5 MiB"] installHandler["388.0 MiB"]
packages["Installed<br/>packages"] packagesMcp["51"] packagesFast["180"] packagesOfficial["119"] packagesXmcp["171"] packagesSkybridge["300"] packagesHandler["130"]
views["Views"] viewsMcp["✅"] viewsFast["✅"] viewsOfficial["◐ Extension"] viewsXmcp["✅"] viewsSkybridge["✅"] viewsHandler["❌"]
nativeViews["Native Views<br/>on MCP 2026"] nativeViewsMcp["✅"] nativeViewsFast["✅"] nativeViewsOfficial["❌"] nativeViewsXmcp["❌"] nativeViewsSkybridge["❌"] nativeViewsHandler["❌"]
oauth["One-line<br/>OAuth adapters"] oauthMcp["✅"] oauthFast["◐ Provider/proxy"] oauthOfficial["◐ Primitives"] oauthXmcp["✅"] oauthSkybridge["✅"] oauthHandler["❌"]
protocol["MCP 2026<br/>protocol"] protocolMcp["✅"] protocolFast["✅"] protocolOfficial["✅"] protocolXmcp["❌"] protocolSkybridge["❌"] protocolHandler["❌"]
screenshot["Built-in View<br/>screenshot CLI"] screenshotMcp["✅"] screenshotFast["❌"] screenshotOfficial["❌"] screenshotXmcp["❌"] screenshotSkybridge["❌"] screenshotHandler["❌"]
tunnel["Built-in<br/>tunneling"] tunnelMcp["✅"] tunnelFast["❌"] tunnelOfficial["❌"] tunnelXmcp["❌"] tunnelSkybridge["✅"] tunnelHandler["❌"What people ask about mcp-use
What is mcp-use/mcp-use?
+
mcp-use/mcp-use is mcp servers for the Claude AI ecosystem. The fullstack MCP framework to develop MCP Apps for ChatGPT / Claude & MCP Servers for AI Agents. It has 10.6k GitHub stars and its last recorded update is dated 2026-09-10.
How do I install mcp-use?
+
You can install mcp-use by cloning the repository (https://github.com/mcp-use/mcp-use) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is mcp-use/mcp-use safe to use?
+
Our security agent has analyzed mcp-use/mcp-use and assigned a Trust Score of 100/100 (tier: Verified). See the full breakdown of passed checks and flags on this page.
Who maintains mcp-use/mcp-use?
+
mcp-use/mcp-use is maintained by mcp-use. The last recorded GitHub activity is dated 2026-09-10, with 50 open issues.
Are there alternatives to mcp-use?
+
Yes. On ClaudeWave you can browse similar mcp servers at /categories/mcp, sorted by popularity or recent activity.
Deploy mcp-use to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/mcp-use-mcp-use)<a href="https://claudewave.com/repo/mcp-use-mcp-use"><img src="https://claudewave.com/api/badge/mcp-use-mcp-use" alt="Featured on ClaudeWave: mcp-use/mcp-use" width="320" height="64" /></a>More MCP Servers
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
An open-source AI agent that brings the power of Gemini directly into your terminal.
Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface
The fastest path to AI-powered full stack observability, even for lean teams.
🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!