Internal Knowledge Base
The internal Knowledge Base is a tool that allows users to upload and manage their own data. It provides a interface for creating collections, uploading files, and managing the data within those collections. The main goal of the internal Knowledge Base is to enable users to easily provide agents with access to additional information. Agents will be able to search and retrieve data from the internal Knowledge Base through Agentic Retrieval Augmented Generation (RAG).
The Knowledge Base includes four main sections:
- Files - This section displays the list of all files that have been uploaded to the Knowledge Base in a table view. Each row represents a single file with the file name, name of the collection it belongs to and buttons to Download and Delete the file. The right collapsable panel provides an area to drag and drop files for upload. Users can also select a collection from the list to which the file will be uploaded together with parameters on how to split file.
- Data - This tab allows users to browse through all data that has been uploaded to a given collection. All elements will be displayed in a paginated form.
- Search - Search will allow user to perform a semantic search on a given collection and simulate the process that agent will be using during the retrieval of data.
- Collections - This section allows users to create and manage collections. Users can create new collections, view existing collections, and delete collections.
To use the internal Knowledge Base, users can create collections and upload files into those collections. The internal Knowledge Base supports various file formats, including text files, PDFs, and presentations. Before the files are uploaded, user needs to enable at least one embedding model. The process do enable embedding models is detailed in Enable Embedding Model section.
