Adam Wanninger's MCP Servers
A collection of MCP servers built and run by Adam Wanninger.
The Model Context Protocol (MCP) lets an AI assistant such as Claude call tools that live on another server. Each server listed here gives your assistant a small set of tools. Add the server's address to your MCP client, sign in with your Google account, and your assistant can use them.
Servers
QueensCoach
Live bus and light rail data for the Charlotte Area Transit System (CATS): where a vehicle is right now, and when the next one reaches a stop.
find_vehiclefinds one bus or train, or every vehicle on a route.list_vehiclestakes a snapshot of every vehicle's position.get_arrivalspredicts arrival times at a stop.
- Address
https://queenscoach.adamwanninger.com/mcp- Source
- github.com/ajwann/queenscoach, under the MIT license. It's also on PyPI, so you can run it on your own computer without signing in.
Not affiliated with CATS or the City of Charlotte. The data comes from CATS's public transit feeds.
Connecting
Claude: open Settings, then Connectors, choose Add custom connector, and paste the server's address.
Claude Code: add the server, then run /mcp to sign in.
claude mcp add --transport http queenscoach https://queenscoach.adamwanninger.com/mcpAny other MCP client that supports remote servers with OAuth sign-in works the same way.
Signing in
The first time your client connects, a browser window asks you to sign in with Google. Google tells the server your account ID and email address. The server checks that the email address is verified, then gives your MCP client an access token of its own.
The server never sees your Google password. It gets no access to your Gmail, Drive, contacts, or anything else in your Google account, and it doesn't keep your email address. The privacy policy has the details.
Fine print
These servers are a personal project. They're free to use and come with no guarantee of uptime: any of them may be slow, unavailable, or shut down. Using them means you accept the terms of service.