Amazon Devices Builder Tools (ADBT) for AI accelerates the development of high-quality Fire TV apps. ADBT for AI is a suite of AI-powered development tools for Fire TV, including ADBT Model Context Protocol (MCP) Server and Agent skills. These capabilities make your coding agents such as Claude Code, Cursor, and Kiro aware of specialized Fire TV knowledge and best practices, eliminating guesswork from development and debugging tasks.
Whether you are new to building Fire TV apps or are a seasoned TV app developer, Builder Tools for AI makes your entire development workflow easier and faster. Today, Builder Tools for AI supports feature integration (In-App Purchase specifically) and remote document search and retrieval for both Vega OS and Fire OS. Workflows for jumpstarting your app migration from Fire OS to Vega are also supported for WebView and React Native app types. For devices powered by Vega OS, Builder Tools for AI additionally supports performance debugging, crash debugging, and media player workflows. For developers building on Fire OS, performance debugging, Fire OS SDK upgrades, and additional integrations are coming soon.
Why it matters
Your AI assistants perform best when they have domain-specific knowledge. Out of the box, LLMs bias towards best practices suited for mobile and web app development and often miss the nuances of TV app development: D-pad navigation, focus management, 10-foot UI design, media pipeline integration, or Fire TV’s recommended patterns. Without this context, AI assistants fall back to patterns that do not translate well to the living room and require constant nudging to steer them in the right direction.
That’s the gap we’re closing. With Builder Tools for AI, your coding assistant gains insight into the nuances of building apps for Fire TV, enabling you to build high-quality Fire TV apps faster.
What can I do with Builder Tools for AI?
Builder Tools for AI supports your entire development workflow from onboarding to advanced use cases like feature integration, performance optimization, and crash debugging. A single command installs all required capabilities including MCP and Skills on your machine. You do not need to think about which capability to use; your agent automatically selects the right workflows and skills based on your prompt. The table below highlights a few examples, see the Amazon Devices Builder Tools for AI documentation more details.
Onboarding
Onboard and set up your environment for Vega app development.
Sample prompts:
Help me setup the Vega SDK
Help me setup the Vega app project
Development Tasks
Perform common development tasks using natural language.
Sample prompts:
Help me validate my manifest file
Guide me to implement focus management for my Vega app
Implement Carousel for my app
App Migration (Fire OS to Vega) — Beta
Port your Fire OS WebView and RN apps to Vega apps while maintaining visual and functional consistency.
Sample prompts:
Port my Fire OS Web app to a Vega Web app
Convert my Fire OS RN app to a Vega RN App
Feature Integrations
Integrate and test Amazon Appstore In-App Purchasing (IAP) SDK in Vega or Fire OS apps.
Sample prompts:
Help me integrate IAP in my Vega app
Help me integrate IAP in my Fire OS app
Performance
Diagnose and fix performance issues.
Sample prompts:
Help me fix frame drops and jank in my Vega app
Can you help me minimize unnecessary re-renders in my app?
Crash Analysis
Diagnose JavaScript, Native, and Low Memory Killer (LMK) crashes with Automated Crash Report (ACR) analysis.
Sample prompts:
Why did my app crash
Help me analyze this ACR file
Doc Search
Search content from developer.amazon.com.
Sample prompt:
What are the submission requirements for my app?
Media Player
Upgrade shaka player.
Sample prompts:
Implement headless media playback architecture for my Vega app
Update Shaka Player from version x to version y for my Vega app
Does my data remain private?
The Builder Tools for AI do not collect, transmit, or store any of your confidential information, including your code samples, your prompts, your project files, or any other developer data. All interactions between the Builder Tools and your AI assistant happen locally except when the agent needs to find relevant documentation, e.g., fetching public documentation from developer.amazon.com.
Get started today
To get started, make sure you have an MCP-compatible AI Coding assistant such as Claude Code, Cursor, or Kiro. Then run this command from your terminal:
npx -y @amazon-devices/amazon-devices-buildertools-mcp init-context
This downloads, installs, and configures all supported AI capabilities — MCP and Skills. Your setup stays current automatically: each session checks the latest capabilities and prompts you to download any updates, so you are always working with the newest workflows.
We also offer community skills that work across open-source frameworks like React Native. To contribute to these skills and help grow the library, see the Device Agent Skills GitHub repo for more details.
We are committed to helping you build high-quality apps for Amazon devices even faster. This is our first major release, and we’re actively expanding the skills and workflows we support. We are learning what works best for TV app developers alongside you, and your feedback directly shapes what we build next. Visit Amazon Devices Builder Tools for AI to get started.
Last updated: Aug 19, 2026