AI Items Store React
marcelpujol/ai-items-store-react
Assistente IA para gerenciamento de inventário com streaming de agentes
Ver no GitHubAssistente de IA para gestão de inventário de loja com arquitetura de três camadas: frontend em React com UI responsiva, componentes remotos via Module Federation, e backend Node.js integrando Claude 3.5 Sonnet via Amazon Bedrock para execução de tools e operações CRUD em banco de dados.
Instalação
npx convez-ai marcelpujol/ai-items-store-react1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Consultas inteligentes sobre estoque e disponibilidade de produtos
- Automação de operações CRUD em banco de dados através de linguagem natural
- Componentes dinâmicos renderizados conforme demanda do usuário
Veredito

Projeto bem estruturado em monorepo com separação clara de responsabilidades. Ideal para quem quer aprender integração de agentes Claude com aplicações React em produção, embora exija configuração manual de credenciais cloud (Bedrock/OpenAI).
README
A tradução para o português deste README ainda está na fila.
Context
The project consists of three solutions (packages), including two React applications (without any associated framework) and a backend solution implemented in Node.js:
-
Frontend package - Assistant
- Contains the frontend code for the virtual assistant.
- Utilizes React hooks provided by Vercel’s AI SDK UI.
-
Frontend package - Remote Component
- Designed specifically to experiment with remote component exposure using Module Federation.
- Enables components to be dynamically rendered in the assistant based on user demand.
-
Backend package - Node.js Solution
- Utilitzes AI SDK Core server functions.
- Configures the agentstreaming text using the Claude 3.5 Sonnet model from Anthropic via Amazon Bedrock.
- Manages the configuration of tools providing a detailed prompt that the agent can execute based on user queries to the LLM.
- Additionally, it has the routes, controllers and services to interact with the Database to perform the CRUD actions.
Development mode
We are going to launch the three packages locally. All commands must be executed from the root directory.
These commands run each app in development mode, meaning they have hot reload enabled to automatically reflect your changes in the browser.
- Launch assistant (React app) --> hosted at http://localhost:3000
cd assistant
npm run start
- Launch Remote (React app) --> hosted at http://localhost:3001
cd remote
npm run start
- Launch Service (NodeJS app) --> hosted at http://localhost:8080
cd service
npm run dev
Environment variables
In the service solution, you need to configure an environment variables file in its root directory, named .env.development. This file should contain the necessary variables for the server (e.g., port, host, user, database name) as well as the AI provider keys you use, such as Amazon Bedrock or OpenAI.
Here is an example of the content for this file:
#Development environment variables
PORT=8080
DB_HOST=localhost
DB_PORT=5432
DB_USER=testuser
DB_PASS=myPass
DB_NAME=items_database
AWS_ACCESS_KEY_ID='[YOUR BEDROCK ACCESS_KEY HERE]'
AWS_SECRET_ACCESS_KEY='[YOUR BEDROCK SECRECT_ACCESS_KEY HERE]'
AWS_REGION='us-east-1'
OPENAI_API_KEY='[YOUR OPENAI API KEY HERE]'
Make sure to replace the placeholder values with your actual credentials before running the service.