### Install MCP Server via npm

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

Use this command to start the MCP server locally. For a full list of server arguments, consult the --help flag.

```bash
npx @cloudinary/structured-metadata-mcp start --api-key  --api-secret  --cloud-name 
```

```bash
npx @cloudinary/structured-metadata-mcp --help
```

--------------------------------

### Gemini MCP Server Installation

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

Add the CloudinarySMD MCP server using the Gemini CLI.

```bash
gemini mcp add CloudinarySMD -- npx -y @cloudinary/structured-metadata-mcp start --api-key  --api-secret  --cloud-name 
```

--------------------------------

### Build and Run MCP Server from Source

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

These bash commands outline the steps to clone the repository, install dependencies, build the project, and run the MCP server locally.

```bash
# Clone the repository
git clone https://github.com/cloudinary/structured-metadata-mcp.git
cd structured-metadata-mcp

# Install dependencies
npm install

# Build the project
npm run build

# Run locally
npm start
```

--------------------------------

### Run MCP Server with Command-Line Arguments

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Runs the MCP server container providing Cloudinary credentials directly as command-line arguments to the start command. Use this if environment variables are not preferred.

```sh
docker run -d -p 2718:2718 \
  cloudinary-structured-metadata-mcp start --transport sse \
  --cloud-name "<your_cloud_name>" \
  --api-key "<your_api_key>" \
  --api-secret "<your_api_secret>"
```

--------------------------------

### Enable Dynamic Mode

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

Start the MCP server in dynamic mode by passing the `--mode dynamic` flag. This registers meta-tools for progressive tool discovery.

```jsonc
{
  "mcpServers": {
    "CloudinarySMD": {
      "command": "npx",
      "args": ["@cloudinary/structured-metadata-mcp", "start", "--mode", "dynamic"],
      // ... other server arguments
    }
  }
}
```

--------------------------------

### CLOUDINARY_URL Format Example

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

This shows the format for using a single CLOUDINARY_URL environment variable to configure authentication, which is an alternative to setting individual API key, secret, and cloud name variables.

```bash
CLOUDINARY_URL=cloudinary://API_KEY:API_SECRET@CLOUD_NAME
```

--------------------------------

### Claude Code CLI MCP Server Installation

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

Add the CloudinarySMD MCP server using the Claude Code CLI.

```bash
claude mcp add CloudinarySMD -- npx -y @cloudinary/structured-metadata-mcp start --api-key  --api-secret  --cloud-name 
```

--------------------------------

### Build Docker Image from Local Clone

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Builds the Docker image for the MCP server from a local repository clone. Use this command in the root of the project.

```sh
docker build -t cloudinary-structured-metadata-mcp .
```

--------------------------------

### View Live Docker Container Logs

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Follows and displays Docker container logs in real time. Press Ctrl+C to stop following. Replace <your_container_id> with the actual ID.

```sh
docker logs --follow <your_container_id>
```

--------------------------------

### Run MCP Server with Individual Environment Variables

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Runs the MCP server container using individual environment variables for Cloudinary credentials. This is the recommended method for providing credentials.

```sh
docker run -d -p 2718:2718 \
  -e CLOUDINARY_CLOUD_NAME="<your_cloud_name>" \
  -e CLOUDINARY_API_KEY="<your_api_key>" \
  -e CLOUDINARY_API_SECRET="<your_api_secret>" \
  cloudinary-structured-metadata-mcp start --transport sse
```

--------------------------------

### Build Docker Image from GitHub

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Builds the Docker image directly from the GitHub repository without needing a local clone. This is useful for quick builds or CI/CD pipelines.

```sh
docker build -t cloudinary-structured-metadata-mcp https://github.com/cloudinary/structured-metadata-mcp.git
```

--------------------------------

### Run MCP Server with Existing Environment Variables

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Runs the MCP server container passing existing shell environment variables for Cloudinary credentials. This avoids re-typing credentials if they are already set.

```sh
docker run -d -p 2718:2718 \
  -e CLOUDINARY_CLOUD_NAME \
  -e CLOUDINARY_API_KEY \
  -e CLOUDINARY_API_SECRET \
  cloudinary-structured-metadata-mcp start --transport sse
```

--------------------------------

### Run MCP Server with CLOUDINARY_URL Environment Variable

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Runs the MCP server container using the CLOUDINARY_URL environment variable, which consolidates all Cloudinary credentials into a single string. This is a concise way to provide credentials.

```sh
docker run -d -p 2718:2718 \
  -e CLOUDINARY_URL="cloudinary://<your_api_key>:<your_api_secret>@<your_cloud_name>" \
  cloudinary-structured-metadata-mcp start --transport sse
```

--------------------------------

### MCP Server Authentication Configuration

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

This JSON snippet demonstrates how to configure the MCP server's environment variables for authentication, including cloud name, API key, and API secret.

```json
{
  "env": {
    "CLOUDINARY_CLOUD_NAME": "demo",
    "CLOUDINARY_API_KEY": "123456789012345",
    "CLOUDINARY_API_SECRET": "abcdefghijklmnopqrstuvwxyz12"
  }
}
```

--------------------------------

### Enable Debug Logging with CLOUDINARY_DEBUG Environment Variable

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Runs the MCP server container with debug logging enabled by setting the CLOUDINARY_DEBUG environment variable to true. This is an alternative method for enabling detailed logs.

```sh
docker run -d -p 2718:2718 \
  -e CLOUDINARY_URL \
  -e CLOUDINARY_DEBUG=true \
  cloudinary-structured-metadata-mcp start --transport sse
```

--------------------------------

### Run MCP Server with Existing CLOUDINARY_URL

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Runs the MCP server container passing an existing CLOUDINARY_URL environment variable from the shell. This simplifies the command if the variable is already configured.

```sh
docker run -d -p 2718:2718 -e CLOUDINARY_URL cloudinary-structured-metadata-mcp start --transport sse
```

--------------------------------

### Enable Debug Logging with --log-level Flag

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Runs the MCP server container with debug logging enabled using the --log-level flag. This provides more detailed output for troubleshooting.

```sh
docker run -d -p 2718:2718 \
  -e CLOUDINARY_URL \
  cloudinary-structured-metadata-mcp start --transport sse --log-level debug
```

--------------------------------

### View All Docker Container Logs

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Displays all logs generated so far for a specific Docker container. Replace <your_container_id> with the actual ID obtained from 'docker ps'.

```sh
docker logs <your_container_id>
```

--------------------------------

### Enable Dynamic Mode with Scope Filter

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

Combine dynamic mode with scope filtering by specifying the `--scope` argument. This further refines tool accessibility for the LLM.

```jsonc
{
  "mcpServers": {
    "CloudinarySMD": {
      "command": "npx",
      "args": ["@cloudinary/structured-metadata-mcp", "start", "--mode", "dynamic", "--scope", "admin"],
      // ... other server arguments
    }
  }
}
```

--------------------------------

### Cursor MCP Server Configuration

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/README.md

Paste this JSON into Cursor Settings under Tools and Integrations to add the Cloudinary Structured Metadata MCP Server.

```json
{
  "command": "npx",
  "args": [
    "@cloudinary/structured-metadata-mcp",
    "start",
    "--api-key",
    "",
    "--api-secret",
    "",
    "--cloud-name",
    ""
  ]
}
```

--------------------------------

### Stop MCP Docker Container

Source: https://github.com/cloudinary/structured-metadata-mcp/blob/main/DOCKER.md

Stops the MCP server Docker container. This command finds the container ID based on the image name and then stops it.

```sh
docker stop $(docker ps -a -q --filter "ancestor=cloudinary-structured-metadata-mcp")
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.