Use the MediaFlows MCP server (video tutorial)

Overview

This demo uses the MediaFlows MCP server to create a new automation using a natural language prompt. The automation automatically generates a description of an image and saves it to metadata for use as alt text.

Learn how to configure the MediaFlows MCP server.

Video tutorial

Tutorial contents

This tutorial presents the following topics. Click a timestamp to jump to that part of the video.

Introduction to MediaFlows MCP

0:04	Learn how the MediaFlows MCP server creates automated workflows through natural language prompts. In this example, we'll see how a simple prompt generates a complete flow that analyzes images and saves descriptions as alt text metadata.

Setting up the prompt

0:16	The prompt for this automation is: "Create a MediaFlows flow that saves a description of the image to metadata for use as alt text." This clear instruction tells the MCP server exactly what kind of flow to create.

Use the MediaFlows MCP server

0:27	With the MediaFlows MCP server configured and enabled, Cursor can use it to fulfill the objectives of the prompt. It doesn't need to be told explicitly to use the MCP server but infers it from the prompt. After entering the prompt, you see Cursor making requests to use MediaFlows tools to create the automation flow.

Creating the flow structure

1:13	The MCP server creates a flow with multiple blocks: a trigger that monitors for new uploads, an AI block that analyzes the image content, and an update block that saves the generated description to structured metadata.

Testing the automation

3:43	The flow is tested by uploading a test image. Initially, it fails because the product environment doesn't have the Cloudinary AI Content Analysis add-on enabled. Cursor identifies that the Generate Image Caption block requires this add-on and explains how to enable it. Once the add-on is enabled, the automation successfully processes the image, generates a description, and saves it to the metadata field configured for alt text storage.

Viewing the results

6:24	In the Media Library, you can verify that the automation worked by checking the uploaded asset's structured metadata, where the generated description appears in the text label field, ready to be used as alt text.

Keep learning

If you like this, you might also like...

 

 

Check out the Cloudinary Academy for free self-paced Cloudinary courses on a variety of developer or DAM topics, or register for formal instructor-led courses, either virtual or on-site.