Redis Cloud remote MCP server

Connect an MCP client to the Redis Cloud MCP server that Redis hosts.

Redis hosts a Model Context Protocol (MCP) server for Redis Cloud at https://cloud.mcp.redis.io/mcp. Connect an MCP client to it to manage your subscriptions, databases, and Agent Memory stores from an AI assistant. There is nothing to install or run: you add one URL and your Redis Cloud API keys to your client.

Use requests like these:

• "List my subscriptions and their databases"
• "Create a free Essentials database on AWS in us-east-1"
• "Which regions can I use for an Essentials plan on Google Cloud?"
• "Show the metrics for my Agent Memory store"

Before you begin

You need:

• A Redis Cloud account with the Redis Cloud API enabled.
• An API account key and an API user key. See Manage API keys.
• An MCP client that supports Streamable HTTP and custom request headers. Clients that only support OAuth sign-in can't connect.

The server acts as the API user key's user, so it can do only what that user can do in the Redis Cloud console:

• To create, update, or delete resources, use a user key for a user with the owner role.
• If the user key has a Classless Inter-Domain Routing (CIDR) allow list, add the address your MCP client connects from.

To create, update, or delete resources, your client must also support MCP elicitation (form requests). The server uses it to ask you to confirm a change.

Connect your client

The Agents page in the Redis Cloud console has guided setup. Choose your agent, and the page shows the configuration to copy and links to where you create your API keys.

To connect by hand, use the same settings in every client:

Keep your keys in the client's connection settings. Don't put them in prompts or agent instructions.

Codex

Set the REDIS_CLOUD_API_KEY and REDIS_CLOUD_API_SECRET_KEY environment variables to your API account key and API user key. Then add the server to ~/.codex/config.toml:

[mcp_servers.redis-cloud]
url = "https://cloud.mcp.redis.io/mcp"
env_http_headers = { "x-api-key" = "REDIS_CLOUD_API_KEY", "x-api-secret-key" = "REDIS_CLOUD_API_SECRET_KEY" }

Claude Code

claude mcp add --transport http redis-cloud https://cloud.mcp.redis.io/mcp \
  --header "x-api-key: <your-api-account-key>" \
  --header "x-api-secret-key: <your-api-user-key>"

Cursor

Add the server to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "redis-cloud": {
      "url": "https://cloud.mcp.redis.io/mcp",
      "headers": {
        "x-api-key": "<your-api-account-key>",
        "x-api-secret-key": "<your-api-user-key>"
      }
    }
  }
}

VS Code

Set the REDIS_CLOUD_API_KEY and REDIS_CLOUD_API_SECRET_KEY environment variables to your API account key and API user key before you start VS Code. Then add the server to .vscode/mcp.json in your workspace:

{
  "servers": {
    "redis-cloud": {
      "type": "http",
      "url": "https://cloud.mcp.redis.io/mcp",
      "headers": {
        "x-api-key": "${env:REDIS_CLOUD_API_KEY}",
        "x-api-secret-key": "${env:REDIS_CLOUD_API_SECRET_KEY}"
      }
    }
  }
}

To check the connection, ask your assistant to list your Redis Cloud subscriptions.

What you can do

Your client shows two tools, search_tools and call_tool. Your assistant searches for the Redis Cloud tool it needs and then calls it, so your client's tool list stays short. Clients that support MCP Apps also get call_ui_tool, which opens interactive views such as Agent Memory store metrics.

Payment methods show only the card type, the last four digits, and the expiry date.

Some changes finish in the background. For these, the server returns a task, and your assistant can check the task's status.

Confirm changes

The server asks you to confirm before most changes:

If you decline or cancel, nothing changes. The server never retries a change automatically.

Limits

• Each tool call must finish in about 25 seconds. A change that times out might still complete, so check the resource before you try again.
• Lists return every matching resource in your account. They aren't paginated.
• The server limits how many requests it accepts. Over the limit, calls fail with Global rate limit exceeded. Wait, then try again.
• Calls to manage subscriptions and databases also count toward your Redis Cloud API quota.

Troubleshooting