VeilStrat

You can manage Cloudinary media workflows using MCP servers that connect natural language commands from AI applications to Cloudinary’s asset management, environment configuration, metadata, analysis, and workflow automation tools. These MCP servers let you upload, transform, analyze, organize, and automate media tasks through conversational AI, simplifying complex media pipelines and enabling rapid iteration with AI assistants.

How to use

Use an MCP client to connect to remote MCP servers or run local MCP servers to start interacting with Cloudinary’s media capabilities. Begin by choosing the server you need based on the task: asset management, environment configuration, structured metadata, analysis, or MediaFlows. Then execute actions in natural language through your client to upload assets, apply transformations, query metadata, run AI analysis, or build low‑code workflows. Security and authentication steps must be completed so your requests are authorized for the chosen server.

For remote servers, you’ll point your MCP client at the provided URLs. For local servers, you’ll start a local process that exposes an MCP endpoint on your machine. In both cases, you’ll pass your Cloudinary credentials or an authentication token so the server can perform operations on your Cloudinary account.

How to install

Prerequisites: you need Node.js and npm installed on your system. You may also opt to use Docker for containerized deployments if you prefer.

{
  "mcpServers": {
    "cloudinary-asset-mgmt": {
      "command": "npx",
      "args": ["-y", "--package", "@cloudinary/asset-management-mcp", "--", "mcp", "start"],
      "env": {
        "CLOUDINARY_URL": "cloudinary://api_key:api_secret@cloud_name"
      }
    },
    "cloudinary-env-config": {
      "command": "npx",
      "args": ["-y", "--package", "@cloudinary/environment-config-mcp", "--", "mcp", "start"],
      "env": {
        "CLOUDINARY_URL": "cloudinary://api_key:api_secret@cloud_name"
      }
    },
    "cloudinary-smd": {
      "command": "npx",
      "args": ["-y", "--package", "@cloudinary/structured-metadata-mcp", "--", "mcp", "start"],
      "env": {
        "CLOUDINARY_URL": "cloudinary://api_key:api_secret@cloud_name"
      }
    },
    "cloudinary-analysis": {
      "command": "npx",
      "args": ["-y", "--package", "@cloudinary/analysis", "--", "mcp", "start"],
      "env": {
        "CLOUDINARY_URL": "cloudinary://api_key:api_secret@cloud_name"
      }
    }
  }
}

Configuration and connections

You can connect to remote (cloud-hosted) MCP servers or run local MCP servers. Remote servers are ready to use with the URLs shown in the examples. Local servers run on your machine using npm packages and can be started with the exact commands shown in the local configuration examples.

{
  "mcpServers": {
    "cloudinary-asset-mgmt-remote": {
      "url": "https://asset-management.mcp.cloudinary.com/sse",
      "type": "http",
      "args": []
    },
    "cloudinary-env-config-remote": {
      "url": "https://environment-config.mcp.cloudinary.com/sse",
      "type": "http",
      "args": []
    },
    "cloudinary-smd-remote": {
      "url": "https://structured-metadata.mcp.cloudinary.com/sse",
      "type": "http",
      "args": []
    },
    "cloudinary-analysis-remote": {
      "url": "https://analysis.mcp.cloudinary.com/sse",
      "type": "http",
      "args": []
    },
    "mediaflows-remote": {
      "url": "https://mediaflows.mcp.cloudinary.com/v2/mcp",
      "type": "http",
      "args": []
    }
  }
}

Authentication

Authentication for remote servers uses OAuth2 by default, with API keys as an alternative via headers. For local servers, provide credentials through environment variables or command line options as shown in the configuration examples.

{
  "mcpServers": {
    "cloudinary-asset-mgmt-remote": {
      "url": "https://asset-management.mcp.cloudinary.com/sse",
      "headers": {
        "cloudinary-cloud-name": "your_cloud_name",
        "cloudinary-api-key": "your_api_key",
        "cloudinary-api-secret": "your_api_secret"
      }
    }
  }
}

Troubleshooting

If you encounter interruptions in responses, keep queries concise and limit the number of tools invoked in a single request. Ensure your Cloudinary credentials are correctly configured and have the necessary permissions.

Tools and capabilities

Asset Management server capabilities include uploading, transforming, searching, organizing, and generating archives or download links. Environment Config handles upload presets, transformation settings, streaming profiles, and mappings. Structured Metadata lets you create and query metadata fields, define validation rules, and organize field relationships. Analysis provides AI-powered tagging, moderation, and object/scene detection. MediaFlows enables natural language driven workflow automations, querying existing automations, and debugging configurations.

Notes on access and tools

If you need additional Cloudinary tools, you can request features or report issues to help extend MCP server capabilities.

Available tools

upload

Upload assets (images, videos, raw files) to Cloudinary and organize them in folders and tags.

transform

Apply image and video transformations, including resizing, cropping, format changes, and quality adjustments.

search

Perform advanced searches with filters, tags, folders, and asset properties.

configurePresets

Manage upload presets and transformation settings for consistent workflows.

aiAnalysis

Run AI-powered analysis, moderation, tagging, and captioning on assets.

structuredMetadata

Create and manage structured metadata fields and queries for assets.

workflows

Build and manage automated workflows and PowerFlows with natural language.

Cloudinary

Documentation & install

Installation