Cloudinary
Summary
Cloudinary's official MCP server — upload, search, transform, and organise media assets from an AI client.
Features
- Upload, search, and organise media assets
- Apply transformations and generate delivery URLs
- Manage tags, metadata, and folder structure
Installation
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Description
Cloudinary's official asset-management MCP server gives an assistant control of a media library, which is more useful than it sounds once a library has thousands of assets in it.
What it covers
- Upload from a URL or local file, with automatic format and quality optimisation
- Search the asset library by tag, folder, metadata, or content
- Transformations — resize, crop, format-convert, and apply effects, returning the delivery URL
- Organisation — folders, tags, and structured metadata, applied in bulk
- Asset management — rename, move, and delete
Where it earns its keep
Bulk operations. Retagging a few hundred assets, restructuring folders after a rebrand, or generating a consistent set of derived sizes are all tedious through a dashboard and quick when described once.
The transformation tooling is also genuinely useful during development: getting the right Cloudinary URL for a specific crop and format is fiddly, and having the assistant construct and verify it removes a lot of trial and error.
Setup
Requires cloud name, API key, and secret. Runs with npx -y @cloudinary/asset-management-mcp. Cloudinary publishes companion servers for other parts of its platform in the same repository.
Related MCP Servers
Google Drive
by Model Context Protocol
Search and read Google Drive files — Docs, Sheets, and Slides converted to text an assistant can use.
Obsidian
by Cal Deacon
Let an AI client read and search your Obsidian vault — or any folder of Markdown notes.