Labsco
MCP SERVER · OFFICIAL PROJECT

Cloudinary

by cloudinary

Upload, organize, transform and analyze the media in your Cloudinary account from the conversation.

Office Suites & Cloud File StorageOfficial source
Summary
One account, five servers — enable only the part of Cloudinary you touch.

Splitting asset operations, environment config, metadata and analysis apart means the tools you never call do not sit in every prompt. The transformation tools pay off soonest: they build delivery URLs from the same rules file Cloudinary publishes, instead of leaving the model to guess at transformation syntax.

What it is

Five servers over one Cloudinary product environment — Asset Management, Environment Config, Structured Metadata, Analysis and MediaFlows — added individually so you can enable only the part you work in.

What you get
  • Upload and manage images, videos and raw files, with advanced search and filtering, delete and rename, folders and tags
  • Transformation tools that use the transformation rules file to build delivery URLs and create derived assets
  • Product environment entities: upload presets, upload mappings, named transformations, webhook notifications and streaming profiles
  • Structured metadata fields, values and conditional metadata rules
  • AI content analysis — automatic tagging, moderation, safety checks, object detection and recognition
  • MediaFlows automations for media processing and delivery
Requirements

A Cloudinary account. Remote servers authorize over OAuth in the browser, where you also pick the product environment; API-key headers are accepted as an alternative on every server except Analysis. Local servers run on your machine with credentials configured by hand. New configurations use the /mcp endpoint.

Setup effort

One command plus a key — npx -y --package @cloudinary/asset-management-mcp -- mcp start, then supply credentials