About Cloudinary MCP Server
Cloudinary MCP Server exposes selected Cloudinary Upload and Admin API operations to AI assistants. It supports asset uploads, management, search, and usage reports, while noting that not all API methods and parameters are available.
Key features
- 01Upload assets from URLs, paths, or content
- 02Delete and inspect assets
- 03Search assets with query expressions
- 04Retrieve usage reports
- 05Include tags and metadata in asset results
Use cases
- 01Upload files to a Cloudinary cloud
- 02Find assets by tags or public ID
- 03Check storage, bandwidth, and request usage
MCP support
The server exposes selected Cloudinary API operations as tools for MCP clients, which can upload, manage, search, and inspect assets.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y cloudinary-mcp-serverCLOUDINARY_CLOUD_NAMECLOUDINARY_API_KEYSecretCLOUDINARY_API_SECRETSecretLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
5Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| folder | string | No | Optional folder path in Cloudinary |
| publicId | string | No | Optional public ID for the uploaded asset |
| resourceType | enum | No | Type of resource to uploadimage | video | raw | auto |
| source | any | Yes | The source media to upload (URL, file path, base64 content, or binary data) |
| tags | string | No | A string containing Comma-separated list of tags to assign to the asset |
Details
From the project README
Cloudinary MCP Server
A Model Context Protocol server that exposes Cloudinary Upload & Admin API methods as tools by AI assistants. This integration allows AI systems to trigger and interact with your Cloudinary cloud.
How It Works
The MCP server:
- Makes calls on your behalf to the Cloudinary API
- Enables uploading of assets to Cloudinary
- Enables management of assets in your Cloudinary cloud
It relies on the Cloudinary API to perform these actions. Not all methods and parameters are supported. More will be added over time.
Open an issue with a request for specific method if you need it.
Benefits
- Turn your Cloudinary cloud actions into callable tools for AI assistants
- Turn your Cloudinary assets into data for AI assistants
Usage with Claude Desktop
Prerequisites
- NodeJS
- MCP Client (like Claude Desktop App)
- Create & Copy Cloudinary API Key/Secret at: API KEYS
Installation
To use this server with the Claude Desktop app, add the following configuration to the "mcpServers" section of your claude_desktop_config.json:
{
"mcpServers": {
"cloudinary-mcp-server": {
"command": "npx",
"args": ["-y", "cloudinary-mcp-server"],
"env": {
"CLOUDINARY_CLOUD_NAME": "<cloud name>",
"CLOUDINARY_API_KEY": "<api-key>",
"CLOUDINARY_API_SECRET": "<api-secret>"
}
}
}
}
CLOUDINARY_CLOUD_NAME- your cloud nameCLOUDINARY_API_KEY- The API Key for your cloudCLOUDINARY_API_SECRET- The API Secret for your cloud
Tools
The following tools are available:
-
upload
- Description: Upload a file (asset) to Cloudinary
- Parameters:
source: URL, file path, base64 content, or binary data to uploadfolder: Optional folder path in CloudinarypublicId: Optional public ID for the uploaded assetresourceType: Type of resource to upload (image, video, raw, auto)tags: Comma-separated list of tags to assign to the asset
-
delete-asset
- Description: Delete a file (asset) from Cloudinary
- Parameters:
publicId: The public ID of the asset to deleteassetId: The asset ID of the asset to delete
-
get-asset
- Description: Get the details of a specific file (asset)
- Parameters:
assetId: The Cloudinary asset IDpublicId: The public ID of the assetresourceType: Type of asset (image, raw, video)type: Delivery type (upload, private, authenticated, etc.)tags: Whether to include the list of tag namescontext: Whether to include contextual metadatametadata: Whether to include structured metadata
-
find-assets
- Description: Search for existing files (assets) in Cloudinary with a query expression
- Parameters:
expression: Search expression (e.g. 'tags=cat' or 'public_id:folder/*')resourceType: Resource type (image, video, raw)maxResults: Maximum number of results (1-500)nextCursor: Next cursor for paginationtags: Include tags in the responsecontext: Include context in the response
-
get-usage
- Description: Get a report on the status of your product environment usage, including storage, credits, bandwidth, requests, number of resources, and add-on usage
- Parameters:
date: Optional. The date for the usage report in the format: yyyy-mm-dd. Must be within the last 3 months. Default: the current date
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.