Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/mcp-use/mcp-use/llms.txt

Use this file to discover all available pages before exploring further.

MCP Inspector

The MCP Inspector is an open-source, interactive developer tool for testing and debugging MCP servers. It provides a beautiful, intuitive interface for testing tools, exploring resources, managing prompts, and monitoring server connections - all from your browser.

Key Features

Auto-Mount

Automatically available at /inspector for all mcp-use servers

Multi-Connection

Connect to and manage multiple MCP servers simultaneously

Interactive Testing

Test tools with live execution and real-time results

Real-time Status

Monitor connection states, errors, and server health

OAuth Support

Built-in OAuth flow handling with popup authentication

Widget Support

Full support for MCP-UI and OpenAI Apps SDK widgets

BYOK Chat

Bring Your Own Key chat interface for testing

Saved Tool Calls

Save and replay tool executions for repeated testing

Three Ways to Use

There are three ways to access the MCP Inspector: Visit inspector.mcp-use.com - no installation required! Perfect for quick testing and exploring MCP servers without any setup.

2. Run Locally with NPX

npx @mcp-use/inspector
Opens the inspector in your browser at http://localhost:8080. Great for development when you need local access.
# Auto-connect to a server
npx @mcp-use/inspector --url http://localhost:3000/mcp

# Use custom port
npx @mcp-use/inspector --port 9000

# Combine both
npx @mcp-use/inspector --url http://localhost:3000/mcp --port 9000

3. Auto-mounted with mcp-use Servers

When you create an MCP server with mcp-use, the inspector is automatically available at /inspector:
import { MCPServer } from "mcp-use/server";

const server = new MCPServer({
  name: "my-server",
  version: "1.0.0",
});

server.listen(3000);
// Inspector automatically available at http://localhost:3000/inspector
This is the most convenient option for development as it’s always available alongside your server.

What Can You Test?

The inspector provides comprehensive testing capabilities for all MCP features:
  • Browse all available tools with names and descriptions
  • View detailed JSON schemas for tool parameters
  • Execute tools with custom JSON input
  • View real-time results with syntax highlighting
  • Save tool calls for repeated testing
  • Test MCP-UI and OpenAI Apps SDK widget responses
  • View all available resources with metadata
  • Copy resource URIs for use in applications
  • Check MIME types and resource properties
  • Preview resource content with syntax highlighting
  • Browse available prompts with descriptions
  • View prompt schemas and required arguments
  • Fill in prompt arguments using form interface
  • Render prompts and copy output
  • Handle tool requests that require user input
  • Fill and submit form-mode responses
  • Accept, decline, or cancel requests
  • Support for text, number, boolean, and enum fields
  • Test MCP server with real LLM interactions
  • Bring Your Own Key (BYOK) for OpenAI, Anthropic, etc.
  • Watch tool calls in real-time
  • View detailed execution logs
  • Track how the MCP server processes requests

Privacy & Security

All authentication tokens, credentials, and API keys are stored locally in your browser’s localStorage. Nothing is sent to external servers - everything stays on your device.
The inspector is designed with privacy in mind:
  • Local Storage: All connection data stored in browser
  • No Cloud Sync: Your data never leaves your machine
  • Client-Side Only: All requests go directly from your browser to MCP servers
  • Self-Hosting: Deploy your own instance for full control

Next Steps

Getting Started

Learn how to connect to MCP servers and explore their features

Features

Deep dive into testing tools, resources, prompts, and chat

CLI Usage

Run the inspector locally with command-line options

Self-Hosting

Deploy your own inspector instance with Docker

Build docs developers (and LLMs) love