Uber

Uber is an MCP server that integrates an AI assistant with the Uber Rides API using OAuth 2.0 authentication. It connects developers, automation builders, and end users to Uber services directly inside compatible Model Context Protocol environments like Claude Desktop. Through this server, users can request ride price estimates, book rides, monitor the status of ongoing ride requests, and cancel trips using natural language prompts. The server provides dedicated tools for the OAuth authentication flow, allowing users to generate authorization URLs and store access tokens securely. It supports both a simulated sandbox environment for development without real-world charges and a production mode for live rideshare booking. By delegating trip management to an AI assistant, users can streamline itinerary handling, transport bookings, and ride tracking without manually opening the standard Uber mobile application or dashboard.

Category: CRM, ERP & E-commerce

Tags: booking, rideshare, transportation, uber

Visit Uber

How to install and configure Uber

1. Create an Uber Developer account and register an application in the Uber Developer Dashboard to obtain your UBER_CLIENT_ID and UBER_CLIENT_SECRET. 2. Set your redirect URI to http://localhost:3000/callback and configure the required scopes (profile, request, ride_request). 3. Install the package globally via npm with npm install -g mcp-uber, or run it on demand using npx mcp-uber. 4. Add the server entry to your Claude Desktop configuration file (claude_desktop_config.json): json { "mcpServers": { "uber": { "command": "npx", "args": ["mcp-uber"], "env": { "UBER_CLIENT_ID": "your_client_id", "UBER_CLIENT_SECRET": "your_client_secret", "UBER_REDIRECT_URI": "http://localhost:3000/callback", "UBER_ENVIRONMENT": "sandbox" } } } } 5. Restart Claude Desktop, invoke uber_get_auth_url, complete authorization, and save the token with uber_set_access_token.

What you can do with Uber

• Check ride price estimates between specific pickup and dropoff locations to compare transit costs. * Request an Uber ride directly through conversational prompts without switching to a mobile application. * Monitor live trip progress and status updates for active ride bookings inside your assistant. * Cancel an existing Uber ride request in real time when travel plans or schedules change. * Test rideshare workflows safely using Uber sandbox mode without generating actual driver dispatches or charges.

Key facts

• https://github.com/199-biotechnologies/mcp-uber
• CRM, ERP & E-commerce, Maps, Weather & Local Data
• booking, rideshare, transportation, uber

Part of MCP Servers

How do I install the Uber MCP server?

You can install the Uber MCP server globally on your machine using npm install -g mcp-uber, or run it directly using npx mcp-uber inside your MCP client configuration.

What tools are available in the Uber MCP server?

The server includes tools to generate OAuth authorization URLs, set user access tokens, retrieve ride price estimates, request Uber rides, track active ride request status, and cancel existing rides.

Which MCP clients work with the Uber server?

The server works with any client supporting the Model Context Protocol, including Claude Desktop configured with stdio execution via npm or npx.

Can I test the Uber server without paying for rides?

Yes. By setting the environment variable UBER_ENVIRONMENT to sandbox, the server simulates ride requests and driver assignments without incurring real-world charges or dispatching actual drivers.

What credentials do I need to use the Uber MCP server?

You need an Uber Developer account, an application registered on the Uber Developer Dashboard, a Client ID, a Client Secret, and a configured redirect URI supporting required scopes like request and ride_request.