WebMCP Early Preview: As WebMCP is still under development, Chrome has introduced an Early Preview for developers to test its features and explore how it may work in the future.
Availability :
WebMCP is currently available as an experimental feature behind a flag in Chrome 146. To access and test it, you must enable the appropriate Chrome flag. Refer to Chrome Flags documentation for more information.
Requirements
To use WebMCP, ensure that you have the following:
- Chrome: Version 146.0.7672.0 or later
- Chrome Flag: The WebMCP for Testing flag must be enabled
Setup
- Open Chrome and go to chrome://flags/#enable-webmcp-testing.
- Set the WebMCP for Testing flag to Enabled.
- Relaunch Chrome to apply the changes.
Demo
Install the WebMCP Inspector extension to explore how WebMCP works. The extension allows you to inspect registered functions, execute them manually, and test them with an AI agent.
Get a Gemini API Key: Google AI Studio
Update the API Key in the WebMCP Inspector Extension
Next, scroll down to the Active Provider for AI Chat setting and select Google Gemini from the dropdown menu.
Next, visit your WordPress website and open the WebMCP Inspector extension. Go to the AI Chat tab and use the User Prompt field to ask the AI agent about the available WordPress or WooCommerce tools, or request a specific action. Based on your prompt, the AI agent can use the available tools to perform the requested task on your behalf.
Plugin Log:
Frontend API (Without WebMCP-Enabled Browsers)
1. Retrieve the available tools.
-> window.ibsWebMCPGatewayTools.list()
2. Check the parameters.
