Cloud Assistant: Serverless Conversational AI for AWS

Project Title

Cloud Assistant: An AI-Powered Conversational Interface for AWS

Concise Description

The Cloud Assistant is a web application that provides a conversational interface for interacting with AWS services. Users can use natural language to perform AWS operations, automating tasks and simplifying cloud management. The application is built using a serverless architecture, leveraging AWS Amplify, Amazon Lex, and other AWS services.

Architecture Diagram

Architecture Diagram:

1. Prerequisites:

   • An AWS account with appropriate permissions.
   • Node.js and npm installed.
   • AWS Amplify CLI installed and configured.
   • Git for version control.

2. Clone the Repository:

   git clone https://github.com/GhassanMaq/CloudAssistant
   cd CloudAssistant
   

3. Install Frontend Dependencies:

   npm install
   

4. Set up AWS Amplify:

   • If you haven't already, install the Amplify CLI:

     npm install -g @aws-amplify/cli
     

   • Configure the Amplify CLI:

     amplify configure
     

   • Initialize Amplify in your project:

     amplify init
     

   • Follow the prompts to set up your Amplify environment.

5. Deploy the Backend:

   amplify push
   

6. Configure Environment Variables:

   • If necessary, set any required environment variables for your Lambda functions or other backend resources. This might include API keys, database connection strings, etc.

7. Lex Bot Deployment

   • Import the Lex bot definition file into your AWS Lex console.
   • Build and publish the bot.
   • Configure the necessary Lambda function for intent fulfillment.

8. Frontend Configuration:

   • Ensure that the aws-exports.js file is correctly configured with your AWS resources. This file is typically generated by the amplify init and amplify push commands.

9. Run the Application:

   npm start
   

10. Access the Application:

    • The application should now be running in your browser, typically at http://localhost:3000.

Configuration Requirements and Environment Setup

• AWS Account: You will need an AWS account with sufficient permissions to create and manage the following resources:

  • IAM Users, Groups, and Roles
  • Cognito User Pools
  • Amplify Applications
  • Lex Bots
  • Lambda Functions
  • AppSync APIs
  • DynamoDB Tables

• IAM Permissions: Properly configured IAM roles and policies are crucial for the application to function securely. The following IAM setup is recommended:

  • A group for Frontend Developers with permissions for Cognito, Amplify Console, CloudFormation (read-only), and Lex (for testing).
  • A group for Backend Developers with permissions for Lambda, API Gateway, DynamoDB, AppSync, IAM (read-only), Lex, and S3.
  • A group for DevOps with permissions for CloudFormation, S3, IAM (with caution), and Amplify Console.
  • IAM roles for Lambda functions and Lex to access other AWS services.

• AWS Amplify CLI: The Amplify CLI needs to be configured with your AWS credentials to deploy and manage the application's backend.

• Environment Variables: Set any necessary environment variables as required by your Lambda functions.

Usage Examples with Screenshots

API Reference for your GraphQL Schema

Queries:

• listConversations: Retrieves a list of user conversations.

Mutations:

• createConversation: Creates a new conversation entry.

Information about Lex Intents and Slots

The Amazon Lex bot is designed to understand natural language queries related to AWS operations. Key intents include:

• DescribeS3Buckets: Lists the user's S3 buckets.
• ListEC2Instances: Lists the user's EC2 instances.
• HelpIntent1: Provides information about the bot's capabilities.

Each intent may use slots to capture specific information from the user's input. For example, a future intent to describe a specific EC2 instance might use a slot to capture the Instance ID.

Deployment Instructions

1. Using AWS Amplify Hosting (Recommended):

   • Connect your Git branch to the Amplify Console. This enables full-stack CI/CD, automatically deploying changes to both the frontend and backend when you commit code.

2. Manual Deployments:

   • Use the Amplify CLI to publish your changes:

     amplify publish
     

   • This command builds and publishes both the backend and the frontend of your project.