How to get website mentioned by chatgpt 2026 

Step 1 — Create Your Chatbase Account

Go to chatbase.co and sign up for a free account using your Google or email login. Moreover, the free plan gives you one chatbot with 20 messages per day, which is enough to test your concept fully. Furthermore, no credit card is required to access the free plan and start building immediately. As a result, you can begin building your AI chatbot within the first five minutes of visiting the Chatbase website.

Step 2 — Upload Your Training Data

Training data is the information your chatbot uses to answer questions accurately. Moreover, Chatbase accepts PDF documents, Word files, plain text, website URLs, and Google Docs as training sources. Furthermore, uploading your FAQ document, product catalog, pricing page, and support articles gives the chatbot everything it needs to answer customer queries. As a result, a chatbot trained on your own specific data gives far more accurate, relevant answers than a generic AI assistant.

Good Training Sources for Your AI Chatbot:
- FAQ document (PDF or text)
- Product or service descriptions
- Pricing and plan information
- Return or refund policy
- Support documentation
- Company About Us page
- Blog articles on key topics
- Interview transcripts
      

Step 3 — Configure Your Chatbot Settings

After uploading your data, configure the chatbot’s name, personality, and instruction prompt. Moreover, the system prompt tells the chatbot exactly how to behave, what tone to use, and what topics to avoid. Furthermore, a well-written system prompt is the single most important factor in determining how helpful and on-brand your chatbot feels to users. As a result, spending 20 to 30 minutes crafting a detailed system prompt produces dramatically better chatbot behavior than using the default settings.

You are a helpful customer support assistant for [Your Company Name].
Your job is to answer questions about our products, pricing, and policies
based on the information provided to you.

Always be friendly, concise, and helpful.
If you do not know the answer, say: "I will connect you with our
support team for this question."
Never make up information that is not in your training data.
Always respond in the same language the user writes in.
      

Step 4 — Test Your Chatbot Thoroughly

Use Chatbase’s built-in test chat interface to ask your chatbot the 20 most common questions your customers ask. Moreover, test edge cases like rude questions, off-topic queries, and questions your training data does not cover. Furthermore, note every answer that is incorrect, vague, or off-brand and add more specific training data to address those gaps. As a result, thorough testing before deployment prevents embarrassing chatbot failures in front of real customers.

Step 5 — Deploy to Your Website

Chatbase provides a small JavaScript code snippet that you paste into your website’s header or footer section. Moreover, WordPress users can install the Chatbase WordPress plugin to embed the chatbot in under two minutes. Furthermore, the chatbot automatically appears as a floating chat bubble in the bottom corner of every page on your site. As a result, your AI chatbot goes live for all website visitors within minutes of copying and pasting a single code snippet.

What You Need Before You Start Building

You need Python 3.10 or newer installed on your computer along with a free OpenAI account to get your API key. Moreover, basic Python knowledge covering variables, functions, and loops is sufficient to follow this entire tutorial. Furthermore, a code editor like VS Code makes writing and running the code significantly more comfortable throughout the process. As a result, total setup time before writing your first chatbot line of code is under 20 minutes.

Step 1 — Install Required Libraries

Open your terminal and install the OpenAI Python library using pip. Moreover, the Streamlit library lets you build a clean chat interface without writing any HTML or CSS manually. Furthermore, the python-dotenv library keeps your API key secure and out of your code files during development. As a result, three simple pip install commands give you everything you need to build a complete AI chatbot application.

pip install openai streamlit python-dotenv
      

Step 2 — Set Up Your API Key

Create a file named .env in your project folder and add your OpenAI API key to it. Moreover, never paste your API key directly into your Python code files as this creates a serious security risk. Furthermore, the python-dotenv library loads this file automatically when your application starts running. As a result, your API key stays private and secure throughout the entire development and deployment process.

OPENAI_API_KEY=your_actual_api_key_here
      

Step 3 — Write the Core Chatbot Code

Create a file named app.py and write the main chatbot application code. Moreover, the code below creates a fully functional AI chatbot with a clean web interface using Streamlit in under 50 lines. Furthermore, the conversation history is stored in Streamlit’s session state so the chatbot remembers context throughout each session. As a result, this minimal code produces a professional, working AI chatbot that you can customise indefinitely.

import os
import streamlit as st
from openai import OpenAI
from dotenv import load_dotenv

# Load environment variables
load_dotenv()
client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))

# Page configuration
st.set_page_config(page_title="AI Chatbot", page_icon="💬")
st.title("AI Customer Support Chatbot")

# System prompt - customize this for your use case
SYSTEM_PROMPT = """You are a helpful AI assistant for ValueBharat.
Answer questions clearly, concisely, and helpfully.
If you do not know the answer, say so honestly."""

# Initialize conversation history
if "messages" not in st.session_state:
    st.session_state.messages = [
        {"role": "system", "content": SYSTEM_PROMPT}
    ]

# Display existing messages
for msg in st.session_state.messages:
    if msg["role"] != "system":
        with st.chat_message(msg["role"]):
            st.write(msg["content"])

# Handle new user input
if prompt := st.chat_input("Type your message here..."):
    # Add user message to history
    st.session_state.messages.append(
        {"role": "user", "content": prompt}
    )
    with st.chat_message("user"):
        st.write(prompt)

    # Get AI response
    with st.chat_message("assistant"):
        response = client.chat.completions.create(
            model="gpt-4o",
            messages=st.session_state.messages,
            max_tokens=500,
            temperature=0.7
        )
        ai_reply = response.choices[0].message.content
        st.write(ai_reply)

    # Save AI response to history
    st.session_state.messages.append(
        {"role": "assistant", "content": ai_reply}
    )
      

Step 4 — Run Your Chatbot Locally

Open your terminal, navigate to your project folder, and run the Streamlit command to launch your chatbot. Moreover, Streamlit automatically opens your chatbot in the browser at localhost:8501 within a few seconds. Furthermore, you can test the full conversation experience locally before deploying it to a live server. As a result, the entire local development cycle from writing code to testing in the browser takes under five minutes per iteration.

streamlit run app.py
      

Step 5 — Add Document-Based Knowledge to Your Chatbot

To train your chatbot on your own documents, add a RAG (Retrieval-Augmented Generation) system using LangChain and ChromaDB. Moreover, RAG allows the chatbot to search your uploaded documents before answering, producing far more accurate and specific responses. Furthermore, this approach lets your chatbot answer questions based on your product manuals, support docs, and FAQs rather than general AI knowledge. As a result, a RAG-powered chatbot behaves like a trained expert on your specific business rather than a generic AI assistant.

pip install langchain langchain-openai chromadb pypdf
      

Step 6 — Deploy Your Chatbot Live

Deploy your Python chatbot to Streamlit Cloud for free by pushing your code to a GitHub repository. Moreover, Streamlit Cloud reads your repository, installs dependencies automatically, and gives you a live public URL within minutes. Furthermore, Railway, Render, and Heroku offer alternative free deployment options that support more complex chatbot configurations. As a result, your custom Python AI chatbot can go live on a public URL at zero hosting cost during the initial launch phase.