Searching Your Knowledge Base
Ask questions in plain English and get AI-powered answers from your documents — including deep search for complex queries.
The real power of the knowledge base isn't uploading documents — it's searching them. Instead of scrolling through pages looking for an answer, you ask the AI a question and it finds what you need.
How to search
Navigate to Knowledge Base and use the chat interface. Just type your question the way you'd ask a colleague. The AI searches across all your uploaded documents and returns the most relevant answer.
Here are some effective ways to ask:
- Specific questions — "What's our refund policy for enterprise customers?"
- Summaries — "Summarize the key points from the Q4 sales report"
- Comparisons — "How does our pricing differ between the starter and pro plans?"
- Lookups — "What's the deadline mentioned in the partnership agreement?"
You don't need special syntax or keywords. The AI understands natural language, so just ask like you're talking to someone who's read all your documents.
Understanding your results
When the AI answers a question, it pulls information directly from your uploaded documents. You'll see the response in the chat with references to the source material. This helps you verify where the answer came from and dig deeper if needed.
If the AI can't find a relevant answer in your documents, it'll let you know rather than guessing. This keeps responses grounded and trustworthy.
Deep search
For complex questions that require examining multiple documents or cross-referencing information, Opisense uses deep search (also called agentic retrieval). Deep search goes further than a standard search:
- Examines more sources — It looks across a wider range of documents to find relevant information
- Cross-references — It connects pieces of information from different documents to build a complete answer
- Iterates — It may refine its search based on what it finds, exploring related topics to give you a thorough response
Deep search takes a bit longer than a standard search, but it's worth the wait for questions that span multiple documents or require connecting dots between different sources.
You'll see a visual indicator when deep search is active, showing the AI's progress as it works through your documents. The results include a chain of reasoning so you can see how the AI arrived at its answer.
Tips for better results
- Be specific — "What's the SLA for premium support?" works better than "Tell me about support"
- Include context — "In the onboarding guide, what are the first three steps?" helps narrow the search
- Ask follow-up questions — The chat remembers your conversation, so you can drill down into topics naturally
- Upload more documents — The more content the AI has, the better its answers will be