AI Hub - Knowledge Base
Harness the power of Flexito's OmniAI Knowledge Base to streamline your business operations. By centralising your business information, you enable AI agents to deliver precise and context-rich responses, enhancing customer interactions and boosting efficiency. Discover how to set
Listen to an audio overview
A short, AI-generated podcast-style summary of this article.
Harness the power of Flexito's OmniAI Knowledge Base to streamline your business operations. By centralising your business information, you enable AI agents to deliver precise and context-rich responses, enhancing customer interactions and boosting efficiency. Discover how to set up and leverage this tool to optimise your workflow.
Flexito OmniAI's Knowledge Base serves as a centralised hub for your business information. It allows AI agents and automation flows to access specific data, ensuring they provide users with accurate and context-aware responses.
Structure of the Knowledge Base
The Knowledge Base is made up of two primary components:
Vector Store: This is the main container that organises and groups multiple related files.
Files: These are individual documents, such as PDFs or text files, containing your business information.
Note: All bots within a workspace share Knowledge Bases, promoting efficient reuse of information.
Steps to Create a Knowledge Base
Requirements
Connect your OpenAI API Key by going to:
Workspace → Integrations → Artificial Intelligence → OpenAI
Follow this Step-by-Step Guide to Create a Knowledge Base Access the AI Hub
Open any chatbot and go to AI Hub → Knowledge Base.
Initiate a Vector Store
Select "Create New Vector Store" to begin.
Configuring the Vector Store
Choose a name (e.g., "Business Info").
Set an expiration date if needed (0 means no expiration).
Upload new files or select from existing ones.
Incorporate Files
Select "Files" to add additional documents.
Establish the Knowledge Base
Press the "+" button to create.
Assign a name to your Knowledge Base.
Optionally, add a description.
Choose the Vector Store you created.
Save your settings to finalise.
Utilising the Knowledge Base 1. AI Agents Find it at: AI Hub → AI Agents
→ Model Responses → OpenAI Responses
Setup Instructions:
Activate the "Knowledge Base" option.
Pick the desired Knowledge Base from the dropdown menu.
The AI Agent will use these files to answer user inquiries automatically.
How it Operates: When the AI Agent needs to generate responses about your business, it will automatically reference the files in the selected Knowledge Base.
2. Flow Builder
Find it at: Flow Builder → Actions → Integrations → OpenAI
→ Search Knowledge Base
Settings Available:
Model Response: Choose the response model.
System Message: Set up the AI prompt to guide responses based on business information.
User Input: Typically uses the most recent user text input.
Knowledge Base: Select the appropriate knowledge base.
Max Number of Results: Decide how many results to return (default is 2).
Max Tokens: A minimum of 1,000 tokens is recommended for thorough responses.
Remove Key Values: Optionally remove specific parameters from the payload to stay within the 20,000 character limit for JSON fields.
Benefits of Creating a Knowledge Base 1. Precise and Consistent Responses AI delivers accurate information about your business, products, and services.
It prevents generic or incorrect responses.
Ensures consistency in the information shared.
2. Centralised Management
Update information in one location.
Changes are automatically reflected across all bots in the workspace.
Simplifies maintenance and version control.
3. Time Efficiency
No need to train each bot separately.
Facilitates content reuse across different bots.
Reduces the time needed to set up new agents.
4. Scalability
Easily add new files as your business expands.
Organise information in various Vector Stores by category.
Support multiple knowledge bases for different contexts.
5. Improved User Experience
Provides more relevant and detailed responses.
Decreases the number of unanswered questions.
Offers more efficient and personalised support.
6. Flexible Integration
Utilise in AI Agents for automatic responses.
Integrate into complex flows within the Flow Builder.
Combine with other actions and integrations.
Best Practices
File Organisation
Keep related files together in the same Vector Store.
Use descriptive names for easy identification.
Regularly update information.
Token Configuration
For Knowledge Base searches, use a minimum of 1,000 tokens.
Adjust based on the complexity of responses needed.
Monitoring
Check the status of file processing.
Use "Sync" to refresh the view after making changes.
Test responses after configuration.

