TAIBLES User Guide

Appearance

Introduction: Getting Started ​

Section 2.1: Creating Your First Chatbot ​

Ready to build something amazing? In the next 30 minutes, you'll create a working AI chatbot that answers customer questions automatically. No coding required. No complex setup. Just follow along step-by-step, and you'll have your first AI agent up and running!

What We're Building ​

Imagine this: A customer visits your website at 2 AM and asks "What are your business hours?" Instead of waiting until morning for a response, your AI chatbot answers instantly—friendly, accurate, and helpful.

That's what you're building today. An AI assistant that:

  • Never sleeps or takes breaks
  • Always knows the answers to common questions
  • Responds instantly, 24/7
  • Gets smarter as you teach it more

Sound good? Let's get started!

💡 What You'll Learn

By the end of this section, you'll have:

  • ✅ Created your first table
  • ✅ Added columns for questions and answers
  • ✅ Configured an AI to respond automatically
  • ✅ Tested your chatbot with real questions

Step 1: Create Your Table ​

Every automation starts with a table. Your table is like a workspace where all the magic happens.

Let's create one:

  1. Click "Automations" in the sidebar, then "Create Automation"
  2. In the "Create new Taible" panel, set "Name" to Customer Conversations
  3. Add a "Description": Track conversations with customers and AI responses
  4. Click "Save"

Create new Taible

Set up your workspace

Each row in this table will represent one customer conversation.

What just happened? You created a workspace. Each row in this table will be one customer conversation—one question, one AI answer.

💡 💡 Think of It Like This

Your table is like a guest book at an event. Each row is one person who signed in. Except instead of just names, each row will contain a full conversation—the question they asked and the AI's response!

Another way to think about it: If your table is called "Customer Conversations," then:

  • Each row = one conversation
  • Each column = one piece of that conversation (question, answer, status, etc.)

Simple, right?

Step 2: Add the Incoming Message Column ​

Now we need a place for customer questions to go. This is where you'll type (or customers will send) their questions.

Here's how:

  1. Click "Add Column"
  2. Choose "Text" as the column type, from the "Data & Storage" category
  3. In the "Configure Column" dialog, name the column Customer Question
  4. Click "Save Column"

What just happened? You created a column for incoming messages. This is where customer questions will appear—whether you type them manually (for testing) or they come from a website, email, or WhatsApp later.

💡 The Name You See, and the Name Behind It

The name you type is the label shown in the grid. Behind the scenes TAIBLES also keeps a technical name for the column, which is what you'll reference later when one column needs to read another.

You don't have to manage that yourself—just give the column a clear name.

Step 3: Add the AI Response Column ​

Now for the exciting part—adding the AI that will automatically answer questions! This is where your chatbot comes to life.

Here's how:

  1. Click "Add Column" again
  2. This time, choose "Generate text" from the column types
    • Look for it in the "AI" category
  3. Name the column AI Answer

Add Column

Choose the type of action or data

Text

Store text, names, messages

Generate text

Generate natural language text and responses.

AI
The name you see in the grid

Now configure the AI:

  1. In the "System Prompt", describe how the AI should behave:
You are a helpful customer service assistant.
Answer questions about our products clearly and friendly.
  1. In the "User Prompt", reference your Customer Question column, so each row asks about its own question
  2. Open "Dependencies & Run Order" and make this column depend on Customer Question
    • This tells the AI to wait until there's a question before responding
  3. Pick the AI model you want to use
  4. Click "Save Column"

Configure Column

AI Answer

Customer Question(customer_message)

AI will wait until the customer asks a question before responding

💡 Tip: Be specific about tone, style, and what information to include
What to ask about this specific row

Ready to generate AI responses!

When a customer asks a question, your AI will automatically generate a helpful answer.

What just happened? You taught your AI how to behave! It will:

  • Wait for a customer question
  • Read the question
  • Generate a friendly, helpful answer
  • All automatically!

💡 ✨ The Magic

You just set up automatic intelligence. Every time a new question appears in the "Customer Question" column, your AI will automatically fill in the "AI Answer" column. No clicking, no manual work!

Step 4: Test Your Chatbot! ​

Time to see your agent in action!

Here's how:

  1. Click "Add Row" in the toolbar
  2. In the Customer Question column, type: "What are your business hours?"
  3. Press Enter or click outside the cell
  4. Watch the magic happen! ✨

Customer Conversations

Your first AI agent in action!

Customer Question AI Answer Status

What are your business hours?

We're open Monday through Friday, 9 AM to 6 PM EST. Feel free to reach out anytime and we'll get back to you during business hours!

 ✓ Complete

What are your business hours?

← You just typed this
AI is thinking...
 ⚡ Processing

🎉 You just built your first AI agent!

Watch as the AI automatically generates an answer. It's that simple! You can add more rows to test different questions.

What you'll see:

  1. The Customer Question cell shows your typed question
  2. The AI Answer cell moves to "Queued", then "Running"
  3. After a few seconds, the AI answer appears in place of the status
  4. The cell is now "Completed"—and shows its result rather than a badge

You just built your first AI agent! 🎉

Understanding What Just Happened ​

Let's break down the magic you just witnessed:

The Automatic Flow:

  1. You typed a question → It appeared in the Customer Question column
  2. The AI noticed → Its dependency was satisfied, so the cell was queued
  3. The AI read your instructions → The system prompt told it how to behave
  4. The AI generated an answer → It appeared in the AI Answer column
  5. The cell finished → "Running" gave way to the result ✓

Here's the beautiful part: This happens automatically for EVERY row you add. You don't need to click anything. You don't need to manually trigger the AI. Just add data, and the automation kicks in!

It's like having a tireless assistant who watches your table and jumps into action the moment they see new work to do.

💡 💡 Try These Questions

Add more rows and test these questions:

  • "What's your return policy?"
  • "Do you ship internationally?"
  • "How do I track my order?"
  • "What payment methods do you accept?"

Watch your AI answer each one automatically!

What's Next? Making Your AI Smarter ​

Right now, your AI is pretty clever—it can have conversations and be helpful. But it's giving somewhat generic answers because it doesn't know the specifics of YOUR business yet.

Here's the problem: If a customer asks "What's your return policy?", your AI will make up a reasonable-sounding answer... but it won't be YOUR actual policy!

The solution? Give your AI knowledge! In the next section, you'll learn how to:

  • Upload your FAQ documents, policies, and product guides
  • Give your AI specific knowledge about YOUR business
  • Make responses accurate and personalized to your company
  • Turn your AI from "smart" into "expert"

Think of it like hiring a new employee. Right now, they're friendly and willing to help (great!), but they need training on your company's specifics. That's what we'll do next.

Continue to Section 0.2.2: Adding Knowledge to Your Agent →

Quick Troubleshooting ​

My AI isn't responding

  • Check the dependency on Customer Question under "Dependencies & Run Order"
  • Make sure you filled in a question
  • Wait 10-15 seconds (sometimes it takes a moment)

The answer is too generic

  • That's normal! In Section 2.2, we'll give it specific knowledge
  • For now, it's using general knowledge

I want to edit my AI instructions

  • Open the menu on the AI Answer column header
  • Choose "Configure column"
  • Modify the "System Prompt"
  • Click "Save Column"

🎉 Congratulations! ​

You just built your first working AI agent! Let's celebrate what you accomplished:

  • ✅ Created a table from scratch (your workspace)
  • ✅ Added columns for questions and AI answers
  • ✅ Configured your first AI agent with custom instructions
  • ✅ Saw it answer questions automatically in real-time

This is huge! You now understand the foundation of everything you'll build with TAIBLES.

The Pattern You Just Learned ​

Every automation you create follows this same simple pattern:

  1. Data comes in (questions, orders, emails, forms, etc.)
  2. Actions happen automatically (AI responds, emails send, data enriches, tasks create)
  3. Results appear (answers, confirmations, summaries, reports)

It's always the same: input → automatic processing → output. That's it!

What's Next? ​

Right now, your AI is pretty smart, but it's giving generic answers. In the next section, you'll learn how to give your AI specific knowledge about YOUR business—your products, your policies, your FAQ—so it can give accurate, personalized answers.

Ready to make your agent an expert? Let's go!

Continue to Section 0.2.2: Adding Knowledge to Your Agent →

Built with VitePress

Copyright © 2025 UNIKI GmbH