Knowledge Base
Train your AI on your business. Self-serve setup with automatic synchronization.
3
Input methods
Auto
Sync updates
RAG
Smart retrieval
10MB
Max file size
Website Scraping
Recommended • Automatic updatesEnter your website URL. AI crawls and extracts all relevant content automatically.
How It Works
- Enter your website URL
- AI crawls all accessible pages
- Content is processed and indexed
- Schedule automatic updates
What Gets Extracted
- FAQ and help pages
- Product/service pages
- About and contact info
- Pricing and policies
Automatic Updates
Schedule daily, weekly, or monthly re-scraping. Your AI always has fresh content.
Document Upload
PDFs, text files, and moreUpload product guides, policy documents, FAQs, and training materials.
Supported Formats
- PDF (primary, best extraction)
- TXT, MD, RTF
- CSV, JSON, HTML
- Max 10MB per file
Best For
- Product manuals
- Terms and policies
- Internal knowledge
- Training guides
Direct Context
Always available to AIAdd critical information that your AI should always know. Type or paste directly.
Examples
- • Business hours and contact info
- • Current promotions or announcements
- • Pricing that is not on your website
- • Specific response guidelines
- • Competitor differentiation points
How It Works
Oxaide uses Retrieval-Augmented Generation (RAG) to find the most relevant information for each customer question.
1
Customer asks question
Natural language query is received
2
Semantic search
AI finds most relevant knowledge chunks using vector embeddings
3
Generate response
AI crafts accurate answer using retrieved context + your brand voice
Best Practices
Do
- Start with website scraping
- Schedule automatic updates
- Add direct context for critical info
- Test after adding new content
Avoid
- ✕Duplicate content across sources
- ✕Outdated information
- ✕Vague or ambiguous answers
- ✕Internal-only documents
GDPR/PDPA Compliant
•AES-256 encryption
•High availability
•Business-grade security