Skip to main content
We distinguish between Agents, Knowledge Assets, and Knowledge Sources. This distinction is relevant for further configurations.

​
1. Agents

Agents are found at the highest level of abstraction. These usually refer to an organizational unit in your company or to a particularly large knowledge package. In the Agent Hub in the Genow application, these are arranged horizontally. At the Agent level, there are multiple settings to optimise the data pipeline. Additionally, there are process fit features which can also be configured at this level.

Agent settings

Make use of Agent settings to optimise your Agent.

​
2. Knowledge Assets

Various knowledge packages, so-called Knowledge Assets, can be created for each Agent. These are shown as rectangles in the Genow user interface after an Agent has been selected. User questions are always answered based solely on the knowledge and documents assigned to the selected knowledge assets. In most cases, end users ask questions about all selectable assets. A large number of assets can reduce clarity, but ensures that a subdivision into different knowledge bases becomes apparent. Having a high number of assets increases the cost per request (AI usage costs). It is up to you to define a good middle ground for your Agent. Most Agents have 1-4 assets.

​
3. Knowledge Sources: Foundation of Your Data

Knowledge assets consist of several knowledge sources, i.e., smaller knowledge packages. At the cloud or storage level, knowledge sources can be folders in which files and documents are stored, or the documents themselves. For every knowledge asset, there must be at least one knowledge source! A Knowledge Source is the most direct connection to your company data and the most fundamental level in the knowledge structure of the Genow platform. Think of a Knowledge Source as a precise signpost: it points exactly to a specific location where information is stored, for example, a specific SharePoint folder, a page in Confluence, or a particular Jira project. Two crucial processes take place at this level
1

Synchronization (Sync)

This is where you manually trigger a synchronization. This process loads all current information from the source (e.g., the SharePoint folder) into the Genow platform, making it findable in queries.
2

Permission Assignment

At the Knowledge Source level, you define which users or user groups are allowed to access the data connected here.
Why you can create multiple sources and assign them to an asset? Generally, this allows for extremely flexible and granular control over your data and access permissions.
Why does it also make sense to keep the number of sources low? You can, of course, also select all your relevant directories, folders or documents and assign them to one source and one asset. This sometimes makes the synchronisation and management easier. It can make sense to create multiple sources if you are dealing with a complex topic with many separate sub-topics or complex user authorization concepts. Web searches are also managed at the knowledge source level.

​
More Information About Data Synchronization and Permissions

​
Remarks on Data Synchronization

Find more information about data syncs here.

​
Remarks on User Permissions

Find more information about user permissions here.