This article covers the following:
Overview
The Wingify Model Context Protocol (MCP) server allows you to connect Wingify’s experimentation and insights data directly to Gemini Command Line Interface (CLI). By establishing this connection, you can query the campaign data in your Wingify account directly from your terminal using natural language in Gemini CLI.
This integration connection bridges the gap between AI-driven analysis and your optimization workflow. Instead of navigating through the Wingify dashboard to find specific campaign details, you can ask Gemini to retrieve reports, check the status of feature flags, or summarize recent experiment results with the terminal.
For example, a developer working in the command-line terminal can quickly query: Give me the last paused campaign in my Wingify account. Gemini CLI fetches this data through Wingify’S MCP server in real-time and displays a structured report without the user ever leaving their coding environment.
Key Benefits:
- Contextual Efficiency: Access Wingify data directly within your developer tools and terminal.
- Natural Language Interaction: Query complex campaign data using simple English commands.
- Unified Workflow: Seamlessly switch between code execution and experimentation analysis.
- Rapid Troubleshooting: Quickly identify the state of campaigns or feature flags during development.
The next section explains a step-by-step procedure to integrate the Wingify MCP server with Gemini CLI.
Configure the Wingify MCP Server with Gemini CLI
Follow these steps to integrate the Wingify MCP server with Gemini CLI:
Step 1 - Install and Authenticate Gemini CLI
To connect Wingify’s MCP server with Gemini, you must have the Gemini CLI installed and authenticated in your command line terminal. For more information, refer to Gemini CLI’s installation and authentication steps.
Step 2: Connect the Wingify MCP Server in Gemini CLI
- Open a new terminal session on your system.
- Based on your geographic location, run the following command to establish a connection with the Wingify MCP server:
Region Command Global gemini mcp add --transport http wingify https://mcp.wingify.ai/mcp Europe gemini mcp add --transport http wingify https://mcp-eu.wingify.ai/mcp Asia gemini mcp add --transport http wingify https://mcp-as.wingify.ai/mcp - A confirmation message MCP server "wingify" added to project settings appears upon successful connection.
Step 3: Authenticate with Wingify
- Start Gemini CLI by running gemini command in your terminal.
- Enter the command /mcp auth wingify. Your browser opens the Wingify authorization screen.
- Select your Wingify account and choose the permission level (Browse, Design, or Publish/Admin). You can select from the following permissions to manage how end users interact with Wingify data through the MCP server.
Permission Description Browse Read-only access to Wingify data, session recordings, heatmaps, A/B tests, Split URL, and Multivariate tests. Cannot create or edit. Design Enables creation of new A/B or Split URL tests via the MCP. Publish / Admin Provides full control to create, edit, start, pause, or delete A/B tests and Split URL tests. - Click Approve.
Note: OAuth authentication requires a browser on your local machine. It will not work in headless environments, remote SSH sessions without X11 forwarding, or containerized environments without browser support.
The next section explains how to verify if the Wingify MCP connection is working correctly in Gemini CLI.
Step 4: Verify the Wingify MCP Connection
To ensure Gemini CLI can successfully communicate with your Wingify account, follow these steps:
- In the terminal, enter the command - gemini to enter Gemini CLI
- Run the command /mcp to list all configured servers. Verify that wingify appears in the list of active MCP servers along with its available tools, such as mcp_wingify_get_campaigns_overview_tool and mcp_create_ab_campaign.
Alternatively, you can verify the connection by running a sample prompt in the terminal. For example, Give me the last paused campaign in my Wingify account. Gemini will fetch the data and display a table containing the Campaign ID, Name, Status, and Creation Date directly in the terminal.
Troubleshooting
| Issue | Possible Cause | Recommended Solution |
|---|---|---|
| MCP server connection fails. | Incorrect URL or expired token. | Verify if the MCP URL and token you entered in the command are accurate while connecting to the MCP server as explained in Step 2. Ensure the --transport http flag is included. |
| Wingify MCP tools are not appearing in the Gemini CLI. | Connection timeout or network restrictions. | Check your internet connection and ensure your firewall allows outgoing requests to the Wingify MCP endpoint. |
FAQs
-
Do I need to add the Wingify MCP server every time I open the terminal?
No. Once added, the configuration is saved in your Gemini local settings. You only need to run the gemini command to start interacting with the server. -
What should I do if the MCP server stops responding?
You can try removing and re-adding the Wingify MCP server using the gemini mcp remove wingify and gemini mcp add commands, respectively.
Need more help?
For further assistance or more information, contact Wingify Support.
.