MCP Integration

Objective

Connect your IDE to CloudAgent using the Model Context Protocol (MCP) to build and deploy cloud infrastructure directly from your development environment.

What is MCP? The Model Context Protocol is an open standard that allows AI-powered IDEs like Cursor to connect to external tools and services. CloudAgent's MCP server lets your AI coding assistant deploy infrastructure on your behalf, with built-in guardrails.

Prerequisites

Before setting up MCP, ensure you have:

A CloudAgent account with at least one connected AWS environment
Cursor IDE (opens in a new tab) installed (version 0.43 or later)
A workload configured with your team's security baselines and governance rules

Step 1: Generate Your MCP Configuration

Log in to CloudAgent (opens in a new tab).
Navigate to My Account → Integrations.
Click Generate MCP Config.
Select the Workload you want the IDE to deploy against.
Copy the generated configuration JSON.

⚠️

Security Note: The MCP configuration contains a secure token scoped to your selected workload. Do not share this token or commit it to version control.

Step 2: Configure Cursor

Open Cursor IDE.
Open the Command Palette (Cmd+Shift+P on macOS, Ctrl+Shift+P on Windows/Linux).
Search for "Cursor Settings: Open MCP Config" and select it.
Paste your CloudAgent MCP configuration:

{
  "mcpServers": {
    "cloudagent": {
      "command": "npx",
      "args": ["-y", "@anthropic/cloudagent-mcp-server"],
      "env": {
        "CLOUDAGENT_API_KEY": "your-api-key-here",
        "CLOUDAGENT_WORKLOAD_ID": "your-workload-id"
      }
    }
  }
}

Save the file and restart Cursor.

Step 3: Verify the Connection

Open a new chat in Cursor (Composer or Chat panel).
Type: "List my CloudAgent workloads"
If configured correctly, the AI will respond with your available workloads.

You should see output similar to:

Connected to CloudAgent. Available workloads:
- payments-api (prod)
- user-portal (staging)
- data-pipeline (dev)

Available MCP Capabilities

Once connected, your IDE can perform the following operations through CloudAgent:

Read Operations (No Approval Required)

List workloads and their configurations
View security baselines and governance rules
Check deployment status and history
Query AWS resources within workload scope

Write Operations (Requires Approval)

Deploy infrastructure blueprints
Update workload configurations
Create or modify AWS resources

Guardrails Active: All deployments are validated against your workload's security baselines before execution. If a request violates a security rule, it will be blocked with an explanation.

Troubleshooting

"Connection refused" or "Server not found"

Ensure your API key is valid and hasn't expired
Check that the workload ID matches an existing workload
Verify your network allows outbound HTTPS connections

"Permission denied" on deployment

Your workload may have restrictive security baselines
Contact your security team to review the blocked rule
Check the CloudAgent dashboard for detailed error logs

MCP server not appearing in Cursor

Restart Cursor after saving the MCP configuration
Ensure the JSON syntax is valid (no trailing commas)
Check Cursor's developer console for error messages

Next Steps

Cookbook: Developing in the Corporate Cloud with Guardrails: Learn the full workflow for teams using IDE-based cloud development.
Blueprints and Agents: Create reusable infrastructure templates for your team.
Managing Workloads: Configure security baselines and governance rules.

AI Assistant FAQ