Skip to content

abglnv/mas-aws

Repository files navigation

Quickstart

1. Copy and fill in credentials:

cp .env.example .env
# Edit .env with your AWS credentials and Telegram bot token

2. Start all services:

docker-compose up --build

On first startup the app will automatically:

  • Create PostgreSQL tables and insert sample data (users, products, transactions)
  • Create the Qdrant collection and index sample knowledge base documents
  • Create the DynamoDB table with TTL enabled

API

POST /invoke

// Request
{ "query": "How many completed transactions are there?", "chat_id": "optional-uuid" }

// Response
{
  "response": "There are 7 completed transactions.",
  "sources": ["SQL Database"],
  "token_usage": { "prompt_tokens": 450, "completion_tokens": 80, "total_tokens": 530 }
}

Example Questions

SQL (structured data):

  • "How many users do we have?"
  • "What are the top selling products?"
  • "Show me all pending transactions"

RAG (knowledge base):

  • "What is the return policy?"
  • "How long does shipping take?"
  • "What payment methods are accepted?"

About

Multi-agent RAG + Text-to-SQL system using AWS (DynamoDB, Bedrock), Qdrant, and PostgreSQL - with a Telegram bot interface and streaming API.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages