Set up environment & SDK

This guide shows you how to set up Scalekit in your development environment. You’ll configure your workspace, get API credentials, install the SDK, verify everything works correctly, and optionally set up AI-powered development tools.

Before you begin, create a Scalekit account if you haven’t already. After creating your account, a Scalekit workspace is automatically set up for you with dedicated development and production environments.

Create a Scalekit account

1. Get your API credentials

   Scalekit uses the OAuth 2.0 client credentials flow for secure API authentication.

   Navigate to Dashboard > Developers > Settings > API credentials and copy these values:

   .env

   SCALEKIT_ENVIRONMENT_URL=<your-environment-url> # Example: https://acme.scalekit.dev or https://auth.acme.com (if custom domain is set)
   SCALEKIT_CLIENT_ID=<app-client-id> # Example: skc_1234567890abcdef
   SCALEKIT_CLIENT_SECRET=<app-client-secret> # Example: test_abcdef1234567890

   Your workspace includes two environment URLs:

   Environment URLs

   https://{your-subdomain}.scalekit.dev  (Development)
   https://{your-subdomain}.scalekit.com  (Production)

   View your environment URLs in Dashboard > Developers > Settings.

2. Install and initialize the SDK

   Choose your preferred language and install the Scalekit SDK:

   Node.js

   npm install @scalekit-sdk/node

   Python

   pip install scalekit-sdk-python

   Go

   go get -u github.com/scalekit-inc/scalekit-sdk-go

   Java

   /* Gradle users - add the following to your dependencies in build file */
   implementation "com.scalekit:scalekit-sdk-java:2.0.6"

   <!-- Maven users - add the following to your `pom.xml` -->
   <dependency>
       <groupId>com.scalekit</groupId>
       <artifactId>scalekit-sdk-java</artifactId>
       <version>2.0.6</version>
   </dependency>

   After installation, initialize the SDK with your credentials:

   Node.js

   Initialize SDK

   import { Scalekit } from '@scalekit-sdk/node';
   
   // Initialize the Scalekit client with your credentials
   const scalekit = new Scalekit(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET
   );

   Python

   Initialize SDK

   from scalekit import ScalekitClient
   import os
   
   # Initialize the Scalekit client with your credentials
   scalekit_client = ScalekitClient(
     env_url=os.getenv('SCALEKIT_ENVIRONMENT_URL'),
     client_id=os.getenv('SCALEKIT_CLIENT_ID'),
     client_secret=os.getenv('SCALEKIT_CLIENT_SECRET')
   )

   Go

   Initialize SDK

   import (
     "os"
     "github.com/scalekit-inc/scalekit-sdk-go"
   )
   
   // Initialize the Scalekit client with your credentials
   scalekitClient := scalekit.NewScalekitClient(
     os.Getenv("SCALEKIT_ENVIRONMENT_URL"),
     os.Getenv("SCALEKIT_CLIENT_ID"),
     os.Getenv("SCALEKIT_CLIENT_SECRET"),
   )

   Java

   Initialize SDK

   import com.scalekit.ScalekitClient;
   
   // Initialize the Scalekit client with your credentials
   ScalekitClient scalekitClient = new ScalekitClient(
     System.getenv("SCALEKIT_ENVIRONMENT_URL"),
     System.getenv("SCALEKIT_CLIENT_ID"),
     System.getenv("SCALEKIT_CLIENT_SECRET")
   );

   SDK features

   All official SDKs include automatic retries, error handling, typed models, and auth helper methods to simplify your integration.

3. Verify your setup

   Test your configuration by listing organizations in your workspace. This confirms your credentials work correctly.

   cURL

   Authenticate with client credentials

   # Get an access token
   curl https://<SCALEKIT_ENVIRONMENT_URL>/oauth/token \
     -X POST \
     -H 'Content-Type: application/x-www-form-urlencoded' \
     -d 'client_id=<SCALEKIT_CLIENT_ID>' \
     -d 'client_secret=<SCALEKIT_CLIENT_SECRET>' \
     -d 'grant_type=client_credentials'

   This returns an access token:

   {
     "access_token": "eyJhbGciOiJSUzI1NiIsImInR5cCI6IkpXVCJ9...",
     "token_type": "Bearer",
     "expires_in": 86399,
     "scope": "openid"
   }

   Use the token to access the Scalekit API

   List organizations

   curl -L '<SCALEKIT_ENVIRONMENT_URL>/api/v1/organizations?page_size=5' \
     -H 'Authorization: Bearer <ACCESS_TOKEN>'

   Node.js

   8 collapsed lines
   import { ScalekitClient } from '@scalekit-sdk/node';
   
   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   );
   
   const { organizations } = await scalekit.organization.listOrganization({
     pageSize: 5,
   });
   
   console.log(`Name of the first organization: ${organizations[0].display_name}`);

   Python

   9 collapsed lines
   from scalekit import ScalekitClient
   
   # Initialize the SDK client
   scalekit_client = ScalekitClient(
     '<SCALEKIT_ENVIRONMENT_URL>',
     '<SCALEKIT_CLIENT_ID>',
     '<SCALEKIT_CLIENT_SECRET>'
   )
   
   org_list = scalekit_client.organization.list_organizations(page_size='100')
   
   print(f'Organisation details: {org_list[0]}')

   Go

   5 collapsed lines
   scalekitClient := scalekit.NewScalekitClient(
     <SCALEKIT_ENVIRONMENT_URL>,
     <SCALEKIT_CLIENT_ID>,
     <SCALEKIT_CLIENT_SECRET>
   )
   
   organization, err := scalekitClient.Organization.GetOrganization(
     ctx,
     organizationId
   )

   Java

   7 collapsed lines
   import com.scalekit.ScalekitClient;
   
   ScalekitClient scalekitClient = new ScalekitClient(
     "<SCALEKIT_ENVIRONMENT_URL>",
     "<SCALEKIT_CLIENT_ID>",
     "<SCALEKIT_CLIENT_SECRET>"
   );
   
   ListOrganizationsResponse organizations = scalekitClient.organizations().listOrganizations(5, "");

   If you see organization data, your setup is complete! You’re now ready to implement authentication in your application.

4. Set up Scalekit MCP Server Optional

   Scalekit’s Model Context Protocol (MCP) server connects your AI coding assistants to Scalekit. Manage environments, organizations, users, and authentication through natural language queries in Claude, Cursor, Windsurf, and other MCP-compatible tools.

   The MCP server provides AI assistants with tools for environment management, organization and user management, authentication connection setup, role administration, and admin portal access. It uses OAuth 2.1 authentication to securely connect your AI tools to your Scalekit workspace.

   Building your own MCP server?

   If you’re building your own MCP server and need to add OAuth-based authorization, check out our guide: Add auth to your MCP server.

   Configure your MCP client

   Based on your MCP client, follow the configuration instructions below:

   Claude Desktop
   1. Open the Claude Desktop app, go to Settings, then Developer
   2. Click Edit Config
   3. Open the claude_desktop_config.json file
   4. Copy and paste the server config to your existing file, then save
   5. Restart Claude

   {
     "mcpServers": {
       "scalekit": {
         "command": "npx",
         "args": ["-y", "mcp-remote", "https://mcp.scalekit.com/"]
       }
     }
   }

   Cursor
   1. Open Cursor, go to Settings, then Cursor Settings
   2. Select MCP on the left
   3. Click Add “New Global MCP Server” at the top right
   4. Copy and paste the server config to your existing file, then save
   5. Restart Cursor

   {
     "mcpServers": {
       "scalekit": {
         "command": "npx",
         "args": ["-y", "mcp-remote", "https://mcp.scalekit.com/"]
       }
     }
   }

   Windsurf
   1. Open Windsurf, go to Settings, then Developer
   2. Click Edit Config
   3. Open the windsurf_config.json file
   4. Copy and paste the server config to your existing file, then save
   5. Restart Windsurf

   {
     "mcpServers": {
       "scalekit": {
         "command": "npx",
         "args": ["-y", "mcp-remote", "https://mcp.scalekit.com/"]
       }
     }
   }

   VS Code (1.101+)

   VS Code version 1.101 or greater supports OAuth natively. Configure the MCP server directly without the mcp-remote proxy:

   {
     "servers": {
       "scalekit": {
         "type": "http",
         "url": "https://mcp.scalekit.com/"
       }
     }
   }

   After configuration, your MCP client will initiate an OAuth authorization workflow to securely connect to Scalekit’s MCP server.

   Note

   The Scalekit MCP server source code is available on GitHub. Feel free to explore the code, raise issues, or contribute new tools.

5. Configure code editors for Scalekit documentation

   In-code editor chat features are powered by models that understand your codebase and project context. These models search the web for relevant information to help you. However, they may not always have the latest information. Follow the instructions below to configure your code editors to explicitly index for up-to-date information.

   Set up Cursor

   Play

   To enable Cursor to access up-to-date Scalekit documentation:

   1. Open Cursor settings (Cmd/Ctrl + ,)
   2. Navigate to Indexing & Docs section
   3. Click on Add
   4. Add https://docs.scalekit.com/llms-full.txt to the indexable URLs
   5. Click on Save

   Once configured, use @Scalekit Docs in your chat to ask questions about Scalekit features, APIs, and integration guides. Cursor will search the latest documentation to provide accurate, up-to-date answers.

   Use Windsurf

   

   Windsurf enables @docs mentions within the Cascade chat to search for the best answers to your questions.

   Full Documentation

   @docs:https://docs.scalekit.com/llms-full.txt
   <your question here>

   Costs more tokens.

   Specific Section

   @docs:https://docs.scalekit.com/your-specific-section-or-file
   <your question here>

   Costs less tokens.

   Let AI decide

   @docs:https://docs.scalekit.com/llms.txt
   <your question here>

   Costs tokens as per the model decisions.

   Use AI assistants

   Assistants like Anthropic Claude, Ollama, Google Gemini, Vercel v0, OpenAI’s ChatGPT, or your own models can help you with Scalekit projects.

   Play

   Need help with a specific AI tool?

   Don’t see instructions for your favorite AI assistant? We’d love to add support for more tools! Raise an issue on our GitHub repository and let us know which AI tool you’d like us to document.