Knowledge Base
Last updated
Was this helpful?
A Knowledge Base is the enterprise knowledge encyclopedia for your Agent. You add company documents, FAQs, and web content to the Knowledge Base, and the Agent can answer user questions based on this data.
Without a Knowledge Base, the Agent can only respond using the language model's general knowledge. With a Knowledge Base, it can answer enterprise-specific questions like "What is our company's return policy?"
Document Upload
PDF, Word, Excel, TXT, etc.
Product manuals, internal guidelines, contract templates
FAQ
Question-answer pairs
Frequently asked questions, standard responses
Crawler Import
Automatically scrape content from web pages
Official website information, public data
This is the most frequently asked question:
Data Type
Unstructured (documents, articles, FAQs)
Structured (tables, fields, numbers)
Query Method
Semantic search ("What is the return policy?")
SQL queries ("What was last month's revenue?")
Best For
Knowledge-based questions, policy guidelines, operating instructions
Data queries, statistical analysis, real-time status
Simple rule of thumb: If the answer is in a "document," use a Knowledge Base; if the answer is in a "report," use a Database.
Advanced Note: The Knowledge Base uses RAG (Retrieval-Augmented Generation) technology under the hood. Uploaded documents are split into paragraphs and converted into vectors. When a user asks a question, the system finds the most relevant paragraphs and provides them to the Agent as the basis for its response.
Create a Knowledge Base — Name it and select an Embedding model
Import Data — Upload documents, create FAQs, or use a crawler to scrape content
Search Test — Test the Knowledge Base with sample questions to verify it finds the correct answers
Attach to an Agent — Link this Knowledge Base in the Agent settings
Last updated
Was this helpful?
Was this helpful?