Gemini CLI is an open-source AI agent from Google that brings the power of Gemini directly into your terminal. πŸ’Ž

MCP features supported: Prompts, Tools, Instructions, Dynamic Client Registration

πŸ“– For general Gemini CLI MCP docs, see the official MCP server configuration guide.

Setup

1

πŸ“ Configure the MCP server

Locate your Gemini CLI settings file, usually at ~/.gemini/settings.json, and add the Wonderflow MCP server:

{
  "mcpServers": {
    "wonderflow": {
      "url": "https://catalog-mcp.wonderflow.ai/mcp",
      "oauth": {
        "clientId": "lyLq7SwG6IcGDc9zY52xqTyRcLkUKIo8",
        "redirectUri": "http://localhost:3334/oauth/callback"
      }
    }
  }
}
2

πŸ”‘ Authenticate

Open Gemini CLI and run:

/mcp auth wonderflow

A browser window will open β€” sign in with your Wonderboard credentials. βœ…

3

πŸ”„ Restart and go

Once authenticated, close and reopen Gemini CLI. You’re now ready to use the Wonderflow MCP. πŸŽ‰

4

πŸ§ͺ Test it works

Ask: β€œWhat topics has my catalog discovered recently?”

Approve the tool usage when prompted.

πŸ’‘ You can also configure MCP servers per-project by placing a .gemini/settings.json file in your project root directory.