Pathors home page
English
Search...
⌘K
Dashboard
Dashboard
Search...
Navigation
Tutorials
Connect Knowledge Base to Agent
Guides
API Reference
Get Started
Introduction
Quick Start Guide
Quick Start
Building a Phone Assistant with Vapi
Quick Start Guide to Chat Integration
Quick Start Guide to LINE Bot Integration
Tutorials
Create Conversational Agent
Build a Complex Conversational Agent
Connect Knowledge Base to Agent
Custom Models & Google Calendar Integration
Usage Analytics
On this page
Tutorial Video
Overview
Steps
Related Resources
Tutorials
Connect Knowledge Base to Agent
Learn how to connect a knowledge base to your Pathors agent for smarter answers.
This tutorial will show you how to use the knowledge base feature in Pathors to help your agent answer questions based on your own documents.
Tutorial Video
Overview
Imagine a customer calls your restaurant and asks questions. Instead of writing all answers in the prompt, you can upload documents (like PDFs or TXTs) to the knowledge base. The AI will read these files and answer based on their content.
Steps
Create a Knowledge Base
Go to the knowledge base section in Pathors.
Choose your storage and embedding model (ADA is recommended for best results).
You can adjust chunk size and overlap for more precise answers, but defaults work fine for most cases.
Upload Your Documents
Upload files such as your restaurant’s address, menu, FAQs, etc.
Wait a minute or two for the system to process the files.
Test Search
Try searching for keywords (e.g., “fast food” or “menu”) to see if the knowledge base returns the right information.
Connect to Agent
In your agent’s settings, link the knowledge base.
Start a new chat session to use the latest settings.
Try It Out
Ask your agent questions like “Do you have fast food options?” or “What is the average price per person?”
The agent will answer based on your uploaded documents.
Related Resources
Previous Tutorial: Build a Complex Conversational Agent
Next Tutorial: Custom Models & Google Calendar Integration
Knowledge Base API Reference
Build a Complex Conversational Agent
Custom Models & Google Calendar Integration
Assistant
Responses are generated using AI and may contain mistakes.