# EbbotGPT Knowledge ## EbbotGPT Knowledge In the realm of EbbotGPT, Knowledge refers to where your AI agent sources its information. Under Knowledge, you can craft data sets by adding sources to the data set, such as webscrapes, APIs or uploading files. Information from sources is converted into documents. These documents contain the knowledge in the data sets. {% hint style="info" %} When your **AI agent** receives a question or request, it will searches through the chosen data sets and formulates responses based on the most relevant documents available in that data set. {% endhint %} ### EbbotGPT Knowledge is used in both Chat and Automations The Knowledge part of EbbotGPT is used both when building Chat agents/chatbots and when building workflows in the Automations platform, such as an Email agent. When building workflows using LLM nodes, you choose data sets from your library in the EbbotGPT Knowledge. Note! EbbotGPT Configuration in Chat platform is not relevant when building workflows in Automations. ### EbbotGPT Knowledge Hierarchy * **Data Sets**: These are the primary collections of information you assign to your AI agent. Think of these as the "folder" the AI knows to look in. * **Sources**: A Data Set is built from various sources, the origins of your data. You create these by scraping websites, connecting APIs, or uploading files. * **Documents**: Once a source is added, the information is broken down into individual documents. This is the granular data the AI actually "reads" to find its answers.
EbbotGPT knowledge overview
### Data sets When accessing EbbotGPT Knowledge in the platform, you find the Datasets tab selected by default. In this tab, all data sets are listed. Datasets are compilations of sources that update their documents concurrently. When a dataset is refreshed, it retrieves data from all the sources linked to that specific dataset. When clicking a dataset, you will see the a list of documents associated with the selected dataset. Here, you can search and edit these documents to quickly review the content available. ### Sources Data sets are filled with information by adding sources to the data set. No actual information is stored in the sources, sources are settings and instructions for retrieving information and can therefore be used in different data sets. There are different type of sources available. Under the source tab, you can create and edit your sources settings. Learn more about Source types – click the link below. {% content-ref url="ebbotgpt-knowledge/source-types" %} [source-types](https://docs.ebbot.ai/ebbot-docs/core-capabilities/ebbotgpt/ebbotgpt-knowledge/source-types) {% endcontent-ref %} ### Documents Documents are generated from the data retrieved from sources. Files, websites, APIs and other source types are processed and information is extracted to create documents. Each piece of information is parsed into different fields. An embedder then analyzes these fields to identify relevant documents, enabling the model to formulate responses based on these findings. This structured approach ensures that AI agents can efficiently handle and respond to inquiries by tapping into a rich, organized repository of information, tailored to the specific needs of its requesters/end users. ## Learn more about EbbotGPT Knowledge Explore and learn more related to EbbotGPT Knowledge through the links below. {% content-ref url="ebbotgpt-knowledge/source-types" %} [source-types](https://docs.ebbot.ai/ebbot-docs/core-capabilities/ebbotgpt/ebbotgpt-knowledge/source-types) {% endcontent-ref %} {% content-ref url="ebbotgpt-knowledge/automatic-updates" %} [automatic-updates](https://docs.ebbot.ai/ebbot-docs/core-capabilities/ebbotgpt/ebbotgpt-knowledge/automatic-updates) {% endcontent-ref %} {% content-ref url="ebbotgpt-knowledge/knowledge-pre-processing" %} [knowledge-pre-processing](https://docs.ebbot.ai/ebbot-docs/core-capabilities/ebbotgpt/ebbotgpt-knowledge/knowledge-pre-processing) {% endcontent-ref %} {% content-ref url="ebbotgpt-knowledge/embedder-models" %} [embedder-models](https://docs.ebbot.ai/ebbot-docs/core-capabilities/ebbotgpt/ebbotgpt-knowledge/embedder-models) {% endcontent-ref %} --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.ebbot.ai/ebbot-docs/core-capabilities/ebbotgpt/ebbotgpt-knowledge.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.