CodeAhoy Gen AI Chatbot

This project is a general purpose AI chatbot built using TypeScript and LangChain as an ExpressJS app. The chatbot is designed with a pluggable architecture, allowing you to easily extend and integrate it with various language models and data sources.

Features

Powered by the OpenAI GPT-3 engine, providing rich and interactive conversational experiences.
Uses Redis for efficient session and conversation management.
Adopts clean architecture principles, leading to an easily maintainable and scalable codebase.
In-built RESTful API for chat interactions.
Supports custom prompt templates for versatile conversation flow.

Pre-requisites

Node.js v19 or above
Redis server instance
An API key from OpenAI

Setup

1. Clone the repository

First, clone this repository to your local machine using git clone.

2. Install Dependencies

Navigate to the project root and run npm install to install all necessary dependencies.

3. Define environment variables

Create .env files (.env.development.local, .env.test.local, .env.production.local) at the root of your project and add the following variables:

# PORT
PORT = 3000

# LOG
LOG_FORMAT = dev
LOG_DIR = ../logs

# CORS
ORIGIN = *
CREDENTIALS = true

# OpenAI 
OPENAI_API_KEY = <<YOUR_OPEN_AI_API_KEY_GOES_HERE>>

Replace <<YOUR_OPEN_AI_API_KEY_GOES_HERE>> with your actual OpenAI API key.

4. Run the Application

You can launch the application in development mode by running npm run deploy:dev.

To watch the logs, you can use pm2 logs.

To stop the logs and the application, use pm2 kill.

Testing the API

You can interact with the chatbot using the exposed RESTful API. Here are some examples on how to use it with curl.

1. Create a new conversation

Save the unique id in the response body

curl --location --request POST 'http://localhost:3000/v1/conversations' \
--data ''

2. Send chat messages in the conversation

Use the id to add messages to the conversation

curl --location 'http://localhost:3000/v1/conversations/1ef78d70-a80d-4327-ae33-2c239a815e83/messages' \
--header 'Content-Type: application/json' \
--data '{
    "id": "67357705-7502-4f9e-819a-7656eabff3f5",
    "timezoneOffset": 45,
    "messages": [
        {
            "id": "67357705-7502-4f9e-819a-7656eabff3a6",
            "text": "give me some recommendations for alameda,ca",
            "role": "user"
        }
    ]
}'

3. Get all messages in the conversation

curl --location 'http://localhost:3000/v1/conversations/1ef78d70-a80d-4327-ae33-2c239a815e83'

REST Client (VS Code)

REST Client allows you to send HTTP request and view the response in Visual Studio Code directly.

Modify conversation.http file in src/http folder to your source code.

umermansoor/langchain-typescript-express-starter