Skip to content

Latest commit

 

History

History
96 lines (73 loc) · 5.38 KB

DocumentationStyle1.md

File metadata and controls

96 lines (73 loc) · 5.38 KB
Raw

Documentation: Lambda-Based AI Conversation Handler

Overview

This codebase provides the infrastructure and logic for an AWS Lambda function designed to process new user message events. It leverages AI services to extract requirements from user messages and sends bot responses or error messages to AWS EventBridge for further processing and logging.

Core Components

app.ts

The primary file defining the Lambda handler which integrates various components to:

  1. Process incoming user messages.
  2. Validate and extract requirements using different AI models (OpenAI, Anthropic, Gemini).
  3. Send responses or errors to AWS EventBridge.

Key Functions

  • handler: The entry point for the Lambda function, orchestrating various operations such as validation, extraction, and sending responses.
  • baseHandler: Provides foundational functionality and middleware for the Lambda handler.
  • runConversationLogic: Core logic that handles conversation processing, including invoking requirement extraction and building responses.

Imports and Interactions

  • Event Handling & Validation: Functions and types from lambdaInterface/eventInterface for validating request bodies and defining errors (e.g., BodyJSONParseError, RequestBodyValidationError, getRequestBody).
  • Environment Variables: Validated using functions from lambdaInterface/validateEnvironmentVariables.
  • Secrets Management: Retrieves secrets using functions from secrets/getSecrets.
  • AI Services: Extract requirements with multiple AI models defined in conversationLogic/requirementsExtraction.
  • Response Conversion and EventBridge: Converts bot responses using lambdaInterface/convertBotResponseToFunctionResponse and sends them via lambdaInterface/sendEventBridgeMessage/putEvents.
  • Logging and Middleware: Utilizes @middy/core and @middy/input-output-logger for enhanced logging and middleware capabilities.

Detailed Module Descriptions

secrets/getSecrets.ts

Defines functions for fetching secrets from AWS Secrets Manager:

  • getSecret: Fetch a single secret.
  • getSecrets: Fetch multiple secrets concurrently and return as an object.

lambdaInterface/validateEnvironmentVariables.ts

Handles environment variable validation:

  • ConversationLambdaEnvironmentVariables: Type definition for environment variables.
  • EnvironmentVariableError: Custom error class for missing environment variables.
  • validateEnvironmentVariables: Function to validate and return environment variables.

lambdaInterface/sendEventBridgeMessage.ts

Manages sending events to AWS EventBridge:

  • putEvents: Function that utilizes the AWS SDK to send events to EventBridge.

lambdaInterface/eventInterface.ts

Defines types and utility functions for handling 'newMessage' events:

  • Types like NewUserMessageEvent, ValidRequestBody.
  • Error classes like BodyJSONParseError and RequestBodyValidationError.
  • Functions like getRequestBody, validateRequestBody.

lambdaInterface/convertBotResponseToFunctionResponse.ts

Converts AI bot responses into a function-friendly format:

  • convertBotResponseToFunctionResponse: Processes and structures bot responses.

dateTimeInterface/dateTimeInterface.ts

Utility functions for date and time operations:

  • getCurrentTimezoneAwareDateString
  • getCurrentDayOfWeekString

conversationLogic/getUserRequirementsPrompt/contextBuilder.ts

Builds contexts and validates chat types:

  • Functions like validateChatContext, requirementsContextBuilder.

conversationLogic/getUserRequirementsPrompt/getUserRequirementsPromptBuilder.ts

Generates chat prompts for requirement extraction:

  • requirementsExtractionChatPrompt: Builds prompts from message history, contexts, and response actions.

conversationLogic/getUserRequirementsPrompt/responseActionsBuilder.ts

Creates response actions descriptions:

  • responseActionBuilder: Builds concatenated descriptions.
  • ResponseAction: Type definition used in response action building.

conversationLogic/getUserRequirementsPrompt/systemMessages.ts

Generates system messages for user interaction:

  • Functions like buildSystemMessage1.

conversationLogic/requirementsExtraction.ts

Extracts user requirements using various AI services:

  • requirementsExtraction Functions*: (e.g., requirementsExtractionOpenai, requirementsExtractionAnthropicSonnet).
  • Handles API interactions, logging, and error conversions.

conversationLogic/types.ts

Defines types and enumerations used in chat and bot interactions:

  • Types like Message, ChatMessagesPrompt.
  • Enumerations like BotActionType, BuildFlowMode.

Summary

The codebase is designed to be modular, leveraging AWS services and various AI models to handle user message events robustly. The main functional flows include:

  1. Message Reception: New messages are received and validated.
  2. Requirement Extraction: Employing AI services to understand and extract requirements from messages.
  3. Response Generation: Constructing relevant bot responses.
  4. Event Dispatching: Sending responses to AWS EventBridge for further processing.

By integrating these components, the system successfully manages user interactions, processes complex queries, and provides contextual responses, all while ensuring robust error handling and logging for operational monitoring.# customerDemos