Knowledge Base

Knowledge Bases store and organize information for retrieval-augmented generation (RAG) in your workflows.

Overview#

A Knowledge Base enables your AI workflows to:

• Access your custom data
• Provide accurate, grounded responses
• Answer questions about your documents
• Maintain up-to-date information

Creating a Knowledge Base#

1. Go to Dashboard > Knowledge Base
2. Click Create Knowledge Base
3. Name your knowledge base
4. Add data sources

Data Sources#

Document Upload#

Upload files directly:

• PDF documents
• Text files (.txt, .md)
• Word documents (.docx)

Web Scraping#

Import content from web pages (coming soon).

API Integration#

Connect external data sources via API (coming soon).

How It Works#

1. Ingestion - Documents are processed and chunked
2. Embedding - Text chunks are converted to vectors
3. Storage - Vectors are stored for fast retrieval
4. Retrieval - Queries find relevant chunks by similarity

Using in Workflows#

Add a Knowledge Retrieval Node to query your knowledge base and inject relevant context into LLM prompts.

Best Practices#

1. Keep documents focused on specific topics
2. Update regularly to maintain accuracy
3. Use clear, well-structured documents
4. Monitor retrieval quality and adjust as needed