Scalr's MCP Server: The AI Assistant That Does the Heavy Lifting

What is a Model Context Protocol (MCP) Server?

An MCP Server connects large language models (LLMs) to external data sources and tools so they can act on current information.

Models like Claude and Gemini know a lot, but that knowledge is frozen at training time. The MCP Server gives the model a standard way to reach past that limit.

It's a small standalone service. Its job is to expose live context to the AI application (the MCP Client): up-to-date data, plus access to tools and resources like databases, APIs, or internal files.

Through that interface, the LLM can read information it was never trained on and take actions against the current state of your systems. An AI that only answers from memory cannot do that; one connected to an MCP Server can.

The Scalr MCP Server: AI for Infrastructure as Code

For years, managing cloud environments meant writing CLI commands and working through dense APIs just to see what your infrastructure was doing. Pulling a usage report, auditing tokens, or creating a workspace was manual work that depended on knowing the exact flags and endpoints.

What if you could just ask for it instead?

That is what the Scalr MCP Server does. It connects your AI assistant (like Claude Desktop or VSCode) to your Scalr platform through the Scalr API. It runs locally as a containerized service and turns your plain-English requests into the API calls needed to manage your infrastructure.

---

How the Scalr MCP Server Boosts Efficiency

The Scalr MCP Server takes the manual effort out of daily infrastructure operations. Your team spends less time looking up how to do something, and anyone who can describe what they want can use the Scalr API.

Compliance and Audits

The AI assistant handles governance and security questions that used to mean custom scripts and pulling data from several tools by hand:

• "Review token usage in my account. Show tokens that don't have owners, have never been rotated, and don't have a description."
• "What policy checks have failed the most in the production environment?"
• "List all users and teams that have the admin access policy assigned to them."

Discovery and Management

Instead of searching documentation for the right CLI flag or API endpoint, you can get an end-to-end view of your infrastructure and manage it by asking:

• "Show me the top 10 workspaces with the highest failure rates."
• "Show me all workspaces without drift detection in the production environment."

Less to Memorize for DevOps Teams

With a translation layer between the user and the API, your team no longer has to remember endpoints, query parameters, or filter syntax. The AI turns plain English into the right API calls, so people who aren't fluent in the Scalr API can still get work done with it.

---

Quick Start: Get Running in Minutes

Setting up the Scalr MCP Server is straightforward, requiring only Docker and a Scalr API Token.

1. Get Your API Token: Create a Personal Access Token or Service Account Token in your Scalr account settings with the required permissions.
2. Configure Your AI Client: Add the Scalr configuration to your client's configuration file (e.g., claude_desktop_config.json). This configuration tells your client how to start the local Docker command.

{
  "mcpServers": {
    "scalr": {
      "command": "docker",
      "args": [
        "run",
        "--rm",
        "-i",
        "--pull=always",
        "--env",
        "SCALR_API_TOKEN=your_api_token_here",
        "--env",
        "SCALR_API_URL=https://your-account.scalr.io",
        "scalr/mcp-server:latest"
      ]
    }
  }
}

3. Start Using! After restarting your AI client, you can immediately begin managing and querying your infrastructure using natural language.

---

Learn More: Ready to start talking to your Scalr control plane? Explore the full details in the Scalr MCP Server Documentation.