Skip to main content

​
What is a knowledge base?

A knowledge base is a collection of documents that your agent can search during conversations. When a caller asks a question, the agent automatically retrieves relevant information from the knowledge base and uses it to provide accurate, grounded answers. This is powered by RAG (Retrieval-Augmented Generation) - a technique that combines your data with the LLM’s capabilities for responses that are both intelligent and factually accurate.

​
Why use a knowledge base?

Without a knowledge base, your agent can only rely on its training data and system instructions. With a knowledge base, your agent can:
  • Answer questions about your specific products, services, and policies
  • Reference up-to-date information that the LLM wasn’t trained on
  • Provide accurate pricing, specifications, and other factual data
  • Quote from official documentation and FAQs

​
Supported content types

SourceFormatsDescription
File uploadPDF, TXT, MD, CSV, JSON, HTML, DOC, DOCX, XLS, XLSX, PPTX, RTFUpload documents directly
Web pagesAny public URLScrape and index website content
TextPlain textPaste text content directly

​
How it works

1

Create a knowledge base

Create a named knowledge base to organize related content.
2

Add documents

Upload files, scrape web pages, or paste text content.
3

Processing

Sonara processes your documents - splitting them into chunks, generating embeddings, and indexing them for semantic search.
4

Link to an agent

Connect the knowledge base to one or more agents.
5

Automatic retrieval

During conversations, the agent automatically searches the knowledge base for relevant context before responding.
You can link multiple knowledge bases to a single agent, and the same knowledge base can be shared across multiple agents.