Building with an LLM

Use AI coding assistants to accelerate your Braze integration workflow. Connect your IDE to the Braze Docs MCP server through Context7 and get accurate, up-to-date SDK guidance directly in your development environment.

AI coding assistants can help you write integration code, troubleshoot issues, and explore Braze SDK features—but only if they have the right context. The Braze Docs MCP server provides your AI assistant with direct access to Braze documentation, so it can generate accurate code snippets and answer technical questions based on the latest SDK references.

Connecting to the Braze Docs MCP

Context7 serves as the bridge between your AI assistant and the Braze documentation library. By adding Context7 to your IDE’s MCP configuration, your AI assistant can query the full Braze documentation set and retrieve relevant SDK references, code examples, and integration guides on demand.

Setting up Context7

To connect your AI assistant to the Braze Docs MCP through Context7, add the following configuration to your IDE’s mcp.json file.

• cursor
• claude
• vs code

1
2
3
4
5
6
7
8
{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp@latest"]
    }
  }
}

1
2
3
4
5
6
7
8
{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp@latest"]
    }
  }
}

1
2
3
4
5
6
7
8
{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": ["-y", "@upstash/context7-mcp@latest"]
    }
  }
}

Writing prompts for Braze SDK development

After you set up Context7, include use context7 in your prompts to signal your AI assistant to pull in Braze documentation as context. The following examples show how to write effective prompts for common SDK tasks.

React Native SDK

These prompts demonstrate common integration tasks for the Braze React Native SDK.

Initializing the SDK

1
2
3
Using the Braze React Native SDK, show me how to initialize the SDK 
in my App.tsx with an API key and custom endpoint. Include the 
configuration for automatic session tracking. Use context7.

Logging custom events with properties

1
2
3
4
I need to track user activity in my React Native app using the Braze 
React Native SDK. Show me how to log a custom event called 
"ProductViewed" with properties for product_id, category, and price. 
Use context7.

Setting up push notifications

1
2
3
Using the Braze React Native SDK, walk me through requesting push 
notification permissions on both iOS and Android 13+. Include the 
code for registering the push token with Braze. Use context7.

Handling in-app messages

1
2
3
Show me how to subscribe to in-app messages using the Braze React 
Native SDK, including how to log impressions and button clicks 
programmatically. Use context7.

Web SDK

These prompts demonstrate common integration tasks for the Braze Web SDK.

Initializing the SDK

1
2
3
4
Using the Braze Web SDK, show me how to initialize the SDK with 
braze.initialize(), including the API key, base URL, and options 
for enabling logging and automatic in-app message display. 
Use context7.

Tracking custom events and purchases

1
2
3
4
5
Using the Braze Web SDK, create a JavaScript module that logs a 
custom event called "VideoPlayed" with properties for video_id, 
duration_seconds, and completion_percentage. Also show how to log 
a purchase with product ID, price, currency code, and quantity. 
Use context7.

Registering for web push

1
2
3
4
Using the Braze Web SDK, provide the HTML and JavaScript needed to 
register a user for web push notifications after they click a 
"Subscribe to updates" button. Include the service worker setup. 
Use context7.

Managing user attributes

1
2
3
Using the Braze Web SDK, show me how to set standard user attributes 
(first name, email, country) and custom user attributes (favorite_genre, 
subscription_tier) for the current user. Use context7.

Plain text documentation

You can access the Braze Developer Guide documentation as plain text files optimized for AI tools and LLMs. These files provide Braze documentation in a format that AI assistants can parse and understand without the overhead of HTML rendering.

These files follow the llms.txt standard, an emerging convention for making documentation accessible to AI tools. You can reference these files directly in your prompts or paste their contents into an LLM for context.