💬 AI Clone Chatbot Function – Gemini API (Multi-Language)

This section covers how to build a simple personality-based chatbot function using Google's Gemini API in JavaScript, Python, and PHP.

---

✅ JavaScript (Node.js) – Using @google/genai

📦 Step 1: Install Gemini SDK

npm install @google/genai

🧠 Step 2: Define Chatbot Function

import { GoogleGenAI } from "@google/genai";

const ai = new GoogleGenAI({
    apiKey: "GEMINI_API_KEY", // Replace with your actual API key
});

const chat = ai.chats.create({
    model: "gemini-2.0-flash",
    config: {
        systemInstruction: "You are a clone of person Garvit", // SYSTEM_PROMPT
    },
});

async function get_ai_response(user_input) {
    const response = await chat.sendMessage({
        message: user_input,
    });
    return response.text;
}

// Example call
await get_ai_response("Kya haal hai bhai?"); // "Thik hai bhai, tu suna"

🔍 Explanation

• systemInstruction: Defines bot's personality.
• sendMessage: Sends user input to Gemini.
• response.text: AI-generated reply.

---

✅ Python – Using google-genai

📦 Step 1: Install Gemini SDK

pip install google-genai

🧠 Step 2: Define Chatbot Function

from google import genai
from google.genai import types

GEMINI_API_KEY = "GEMINI_API_KEY"

client = genai.Client(api_key=GEMINI_API_KEY)
chat = client.chats.create(
    model="gemini-2.0-flash",
    config=types.GenerateContentConfig(
        system_instruction="You are a clone of person Garvit"
    ),
)

def get_ai_response(user_input):
    response = chat.send_message(user_input)
    return response.text

# Example call
get_ai_response("Kya haal hai bhai?")  # "Thik hai bhai, tu suna"

🔍 Explanation

• genai.Client: Gemini client object.
• system_instruction: Sets up clone personality.
• send_message: Sends user input, returns reply.

---

✅ PHP – Using Raw curl Request

PHP doesn’t have an official Gemini SDK, but you can use Gemini’s REST API with curl.

🧠 Chatbot Function in PHP

<?php

function get_ai_response(string $userInput): string
{
    $apiKey = 'GEMINI_API_KEY';
    $url = "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent?key=$apiKey";

    $payload = json_encode([
        "system_instruction" => [
            "parts" => [
                ["text" => "You are a clone of person Garvit"]
            ]
        ],
        "contents" => [
            [
                "role" => "user",
                "parts" => [
                    ["text" => $userInput]
                ]
            ]
        ]
    ]);

    $headers = ['Content-Type: application/json'];

    $ch = curl_init($url);
    curl_setopt_array($ch, [
        CURLOPT_RETURNTRANSFER => true,
        CURLOPT_POST => true,
        CURLOPT_HTTPHEADER => $headers,
        CURLOPT_POSTFIELDS => $payload
    ]);

    $response = curl_exec($ch);
    $error = curl_error($ch);
    curl_close($ch);

    if ($error) {
        error_log("Curl Error: " . $error);
        return 'Error contacting AI';
    }

    $responseData = json_decode($response, true);
    return trim($responseData['candidates'][0]['content']['parts'][0]['text'] ?? 'No reply found');
}

// Example call
get_ai_response("Kya haal hai bhai?"); // "Thik hai bhai, tu suna"

🔍 Explanation

• system_instruction: Same as SYSTEM_PROMPT.
• curl: Used to send API request.
• Response is parsed from Gemini’s JSON output.

---

🔚 Summary

Language	SDK/Method	Function Name	Prompt Support
JavaScript	@google/genai	get_ai_response()	✅
Python	google-genai	get_ai_response()	✅
PHP	curl (manual)	get_ai_response()	✅