https://github.com/akhiljw46/kathyhere-backend

A dating AI Chat Bot API
https://github.com/akhiljw46/kathyhere-backend

ai api chatbot expressjs gemini google-gemini javascript nodejs

Last synced: 13 days ago
JSON representation

A dating AI Chat Bot API

• Host: GitHub
• URL: https://github.com/akhiljw46/kathyhere-backend
• Owner: akhiljw46
• License: mit
• Created: 2023-12-16T10:01:41.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2023-12-30T16:42:05.000Z (11 months ago)
• Last Synced: 2024-04-19T01:54:07.678Z (7 months ago)
• Topics: ai, api, chatbot, expressjs, gemini, google-gemini, javascript, nodejs
• Language: JavaScript
• Homepage: https://kathyhere.cyclic.app
• Size: 36.1 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Kathy Here Backend

An AI powered dating chatbot backend powered by [Google Gemini](https://ai.google.dev/)

Created using Express JS

Hosted on : https://kathyhere.cyclic.app

Frontend repository : https://github.com/akhiljw46/kathyhere-frontend

## How to run

### Step 1

Create a `config.env` file in the project folder with the following variables :

`API_KEY` : Google Gemini API key. You can get it [here](https://makersuite.google.com/app/apikey).

`PORT` : The port to run the server. Default port is 3000.

`ENVIRONMENT` : Either `development` or `production`.

`LOG_URL` : The database to save the logs in case of development.

### Step 2

Run the following commands:

```

npm install

npm start

```

## API documentation

Send `POST` request to `{API_URL}/v1/message/{USER_MODEL}`

Where `{API_URL}` is the url of the server and `{USER_MODEL}` is the model to be used, either `kathy` or `tom`.

With payload in the forllowing format:

```

{

    "messages": [

        {

            "id": 1703609459098,

            "isUser": false,

            "messageText": "Message by AI"

        },

        {

            "id": 1703609459099,

            "isUser": true,

            "messageText": "Message by User"

        }

    ]

}

```

Where the `messages` array contains objects with the following properties:

`id` is the unique message id.

`isUser` is a boolean `true` if the message is by the user and `false` if the message is by the AI model.

`messageText` is the message content.

Response will be in the following format:

```

{

    "status": "success",

    "message": "Reply message"

}

```

Where, `status` is either `success` or `failed`

and `message` is the reply message in case of `success` or error message in case of `failed` statuses.