wdio

This MCP server exposes WebdriverIO-based automation tools to a GitHub Copilot Agent operating in Agent Mode. It allows you to describe test scenarios in plain English, and the agent will orchestrate real browser actions through WDIO by calling the exposed tools. Available tools include openBrowser, navigate, clickElement, setValue, getText, isDisplayed, and closeBrowser. The server handles translating high-level scenarios into a sequence of tool invocations and can optionally generate reusable WDIO test code from those scenarios. To use it, run the MCP server locally, connect Copilot Agent in Agent Mode, and enable the tools you want the agent to access. Then you can start describing test flows in natural language, such as navigating to a URL, logging in, interacting with page elements, and validating results, with the agent executing the actions in real time via WDIO.

Prerequisites:

• Node.js (v14+ recommended) and npm/yarn installed
• Access to the MCP server code and repository

Installation steps:

1. Clone the repository git clone https://github.com/your-username/wdio-mcp-server.git cd wdio-mcp-server

2. Install dependencies npm install

3. Configure the MCP server (optional guidance)

   • Ensure mcp.json (or the configuration file used by your setup) references the correct entry point for the Node server (e.g., server.js).
   • If you are using a Docker/VM environment, adjust the docker configuration accordingly.

4. Run the server

   • If using npm scripts, you may run: npm run start
   • Or execute directly: node server.js

5. Verify the server is running

   • Check logs for a ready/listening message and ensure tools are exposed as described in the docs.

6. Connect Copilot Agent in Agent Mode and configure tools

   • In Copilot Agent settings, enable Agent Mode and select the tools exposed by this MCP server (openBrowser, navigate, clickElement, setValue, getText, isDisplayed, closeBrowser).

Tips and troubleshooting:

• If tools are not appearing in Copilot Agent, ensure the MCP server has fully started and that the exposed tool definitions align with what the agent expects.
• Some WDIO actions may require explicit waits or selectors; consider using isDisplayed and getText to validate element readiness before interacting.
• If you modify tools, restart the MCP server to refresh the exposed capabilities.
• Environment variables can be added under the mcpServers.<name>.env section if your WDIO setup requires specific configurations (e.g., browser capabilities, Selenium/WDIO services).
• For local development, ensure a compatible browser driver is installed (e.g., Gecko or Chrome driver) and that WDIO is configured to use it.