This use case showcases how we used Zammo Agents to quickly build and deploy a chatbot powered by Retrieval-Augmented Generation (RAG). The goal was to create a bot that could accurately answer questions based on the content of our website—and cite its sources.


Overview

Using only natural language prompts in the Zammo Agent Designer, we built a fully functional chatbot that pulls answers from an indexed knowledge base tied to our live website content. This example demonstrates how easily organizations can create RAG-powered experiences without writing code or managing complex integrations.


What We Did

✅ Indexed Website Content

We began by creating a Zammo Knowledge Base and indexing content from the Zammo homepage. The indexed content becomes the foundation from which the bot draws relevant answers.

Zammo homepage

Zammo homepage

Zammo knowledge base containing content indexed from the homepage

Zammo knowledge base containing content indexed from the homepage

✅ Connected Content via a RAG Project

Next, we created a RAG Project in Zammo. This step links the knowledge base to retrieval logic that enables the bot to generate context-aware answers and include source citations.

✅ Created the Bot with Natural Language

Using the Zammo Agent Designer, we launched the bot creation process by entering a simple instruction:

“Use my RAG Project to answer customer questions.”

Zammo automatically identified our RAG Project and offered to incorporate additional options such as routing rules, the Agent’s purpose, or a custom welcome message—either during creation or post-launch.