Cloudinary AI Power Start — One Prompt to Get Started

Documentation Index

Fetch the complete documentation index at: https://cloudinary.com/documentation/llms.txt

Use this file to discover all available pages before exploring further.

Image & Video APIs

Last updated: Jul-09-2026

Copy a setup prompt from this page to configure Cloudinary in your active project. The prompts work with Cursor, Claude, Antigravity, GitHub Copilot, and other AI coding assistants.

As you explore this page, you'll learn what the setup prompt installs, what each tool provides, and which recommended tools can further enhance your workflow.

After you run the setup prompt in your IDE, your AI coding assistant will suggest next steps, including prompts for adding Cloudinary image and video features to your project. You can follow those suggestions or return to this page for additional prompts to implement Cloudinary image and video use cases at scale.

On this page:

Start with one prompt
What happens after you paste it
Get the most from your AI coding assistant
Then build something with it
Keep going

Start with one prompt

Open your new or existing project in any AI-powered IDE, then copy and paste the prompt:

cloudinary-get-started-prompt.md View Full Prompt

# Use these instructions to get started with Cloudinary in this directory
This prompt helps a user set up Cloudinary in a new or existing project, or when validating an existing Cloudinary integration. Use to:
– Get started with Cloudinary or CLD
– Configure Cloudinary in a codebase
– Set up Cloudinary in an IDE or agent environment
– Install or configure the right Cloudinary SDK for the detected stack
...

What happens after you paste it

The following five stages run in order. Your agent pauses for confirmation at each gate. Nothing runs without your say-so. Anything already set up in your project or globally in your IDE is skipped.

AI tooling

Installs Cloudinary MCP servers and agentic skills in your IDE.

Framework detection

Confirms your detected stack and how you deliver media.

SDK & environment

Installs the right SDK and helps you set up your .env.

Credentials

Walks you through adding your keys safely.

Verification

Validates your setup with an upload preset, transformation, and optimization check.

Get the most from your AI coding assistant

The Get Started prompt configures your AI coding assistant to work with Cloudinary by setting up MCP servers and agentic skills. Here's what gets set up for you, along with additional tools and tips to help you get the most out of it.

MCP servers

✓ Configured for you

Your AI coding assistant interacts with Cloudinary via MCP servers. Ask it to upload and organize assets, create reusable transformations, and more.

@cloudinary/asset-management @cloudinary/environment-config

You may need to restart your IDE and start the MCP servers. If you're not sure, ask your AI assistant.

MCP servers docs →

Agentic Skills

✓ Installed for you

Your AI coding assistant uses Cloudinary skills for accurate coding & guidance. Basic skills include:

/cloudinary-transformations helps build and troubleshoot delivery URLs.

/cloudinary-docs answers questions using the Cloudinary documentation.

Skills docs →

VS Code extension

Recommended

Install the extension in VSCode, Cursor, or any other VS Code-based IDE to browse, upload, preview, and grab Cloudinary image and video URLs or public IDs without leaving your editor.

**To install:** Open the Extensions panel, search for Cloudinary, and select Install. Your Media Library appears in the sidebar.

Extension docs →

Start prompting

Try it now

With the MCP servers and skills available, describe what you want in plain language. Your AI coding assistant can work with your Cloudinary assets and generate the code you need.

"Create an optimized gallery on this page using Cloudinary images." More example prompts →

For the full guide, see Cloudinary AI agent tools and MCP servers.

Then build something with it

The Get Started prompt ends with a few example prompts to help you build common Cloudinary features. If you're looking for more, copy one of these prompts and paste it into your AI coding assistant.

Important

These prompts work best after you've run the Get Started prompt, which configures your Cloudinary environment for the project.

If you don't run the Get Started prompt, make sure your project is configured with a Cloudinary SDK and make your Cloudinary credentials available to the LLM (e.g., in a local .env file). Never expose API secrets in client-side code.

Social