AI-Wiki

​

What is a Knowledge Source?

A knowledge source is a single valuable knowledge entry in the AI-Wiki. It can be created in different ways:

• as a manually created article
• as content from an externally connected source, for instance, via a website connector
• as an uploaded file through the Files-connector

Manually created articles are suited for content that should be consciously formulated and technically reviewed. This includes operating instructions, frequently asked questions, process descriptions, internal guidelines, or binding formulations. Externally connected sources are suitable when already existing content should be utilized regularly. These can, for example, be help centers, documentation pages, or public information pages. Files are suitable for existing documents that should be centrally provided and made accessible for AI-supported functions, such as PDF documents, operating instructions, or technical documents.

​

Why Structure is Important

Knowledge in the AI-Wiki is not only stored but also cataloged. For this purpose, groups and classifications are used. Groups form the thematic structure. They aid in making content discoverable and bundling technically related information. Classifications describe the type of content. This distinction is important because different types of content are used differently:

• An FAQ answers a specific recurring question.
• An operating instruction describes a procedural sequence.
• A document contains background information or binding regulations.
• News informs about current changes and appears on the homepage.

If these types of content are cleanly separated, the knowledge stock remains easily verifiable. At the same time, AI functions can better classify the content.

​

Website-Connector

The Website-Connector connects an external website to the AI-Wiki. Path in Enneo: Tools → AI-Wiki → Sources During the crawl, Enneo reads the linked website and adopts suitable content as knowledge entries. These then appear in the knowledge structure. A connector is controlled via several settings:

• Source URL
• Include paths
• Exclude paths
• Maximum page number
• Frequency for re-crawling

The source URL defines which website is linked. Include paths determine which areas of the website should be taken into account. Exclude paths exclude areas that are not relevant. These can be, for example, login pages, privacy policy pages, impressum, blog areas, or technical survey pages. Examples:

• /help/* includes contents in the help area.
• /faq/* includes FAQ pages.
• /login/*, /privacy or /impressum can be excluded.

The maximum page number limits the scope of the adopted content. Crawling frequency determines how regularly Enneo re-reads the source. The connector technically provides content. However, it doesn’t evaluate whether this content is technically suitable. This examination remains part of the editorial responsibility.

​

Crawling and Up-To-Dateness

During the crawl, content is read from the external source and made available in Enneo. The connector displays a status, for example, running, completed, or faulty. Re-crawling updates the adopted content based on the current connector configuration. This can incorporate changes on the external website into Enneo. If include or exclude paths, the source URL, or other connector settings are changed, the source could be crawled or processed again so that the changes take effect. At the same time, the quality of the outcome depends on the external source. If URL structures, navigation, or website content change, the knowledge structure in Enneo can also change. For stable results, connectors should be configured as precisely as possible. A narrowly limited source is easier to check and usually delivers better results than a very broadly linked website.

​

Files as Knowledge Sources

In addition to manually maintained articles and linked websites, files can also be used as knowledge sources in the AI-Wiki. Path in Enneo: Tools → AI-Wiki → Sources The Files-Connector is used to centrally provide technically relevant documents and to make them accessible for AI-supported functions. Uploaded files are processed, transformed into knowledge entries and can then be considered in suggested responses, chat contexts, or AI agents, provided they are approved for this. Files can be uploaded using drag-and-drop, structured in folders, managed, and if necessary, checked in the preview. A clear folder structure helps maintain comprehensible larger knowledge stocks and later check them specifically. Files should not be used as unchecked storage. What’s critical is that the content, file name, and storage location are technically clear. If the content of a file changes, the knowledge source must be reprocessed so that the updated content becomes effective in the system. The re-indexing ensures that not only the file itself, but also the search and AI contexts derived from it, are up to date. As with website connectors, the connector technically provides content. The technical responsibility for accuracy, timeliness, approval, and structure remains with the responsible users.

​

Manage, Edit, and Reprocess Files

Inside the Files-Connector, files and folders can be managed. Depending on authorization, users can upload files, move them, check them in the preview, replace, edit, or archive them. For files, the following actions are available, among others:

• Upload file
• Open file in preview
• Replace file
• Edit file, if the format supports this
• Re-index file
• Reset file to the original state
• Archive or restore file

If a file has been edited directly in Enneo, this could be visible through an appropriate marking. This marker helps to recognize that the currently used content may not exactly match the originally uploaded document.

​

Refresh

With Refresh, a file is reprocessed. This action is useful when content for knowledge search or AI-supported functions should be updated. The file is then prepared again for search and AI contexts. This is relevant, for example, when a file was replaced or when the processing should be deliberately triggered again.

​

Reset to Original

With Reset to Original, a manually edited file is reset to the originally uploaded condition. This action is useful when manual changes should be discarded or when the original content of the file should be used as a basis again.

​

Files and Media in ArticlesArticles can be supplemented by uploaded media. This includes images or videos, for example.

Media is useful when it makes a technical statement clearer. This applies to form examples, process presentations or screenshots of certain states. It is important to note that media should supplement the text, not replace it. For AI-supported functions, the written content remains particularly relevant. Text can be precisely searched, processed and adopted into response contexts. Therefore, critical information should always be included in the article text and not contained solely in a screenshot or video.

​

Visibility for AI Functions

The setting ‘Make Knowledge Source Publicly Available’ controls whether a knowledge source can be used for AI-supported functions. If a knowledge source is released, it can be considered, for example, in response suggestions, chat contexts or AI agents. If it is not released, it remains primarily part of the internal documentation in the AI wiki. This setting is technically relevant. A released knowledge source can influence the results of AI functions. Therefore, only contents that are checked, current, and clearly formulated should be released. Contents that are confidential, outdated, incomplete, or technically ambiguous should not be released.

​

Use in Ticket Context

Knowledge sources can be used in the ticket context if they are released for AI functions and technically relevant. Enneo can draw on suitable content from the AI wiki to support response suggestions, chat responses, or AI agents. Furthermore, relevant knowledge content can become visible in the work context of a ticket, enabling users to find suitable information more quickly. Whether a knowledge source is actually considered depends, among other things, on content, release, currentness, and technical relevance.

​

Impact on AI Agents

AI agents can use knowledge sources as context if they are accessible and technically relevant. The quality of the agent results strongly depends on the quality of the knowledge stock. Ambiguous formulations, contradictory articles or untested connector contents can lead to inaccurate results. Therefore, the AI wiki is more than just a repository for information. It controls what knowledge is available to the AI and how this knowledge is technically classified. A well-maintained AI wiki improves the traceability and stability of AI-supported processing. In contrast, a blurred knowledge base can lead to inconsistent responses and misinterpretations.

​

Maintenance and Responsibility

A good knowledge base is not necessarily large but reliable. For manual articles, the following points should be checked in particular:

• Is the content technically correct?
• Is the article clearly formulated?
• Does the classification fit?
• Is the group selected sensibly?
• May the content be used for AI functions?

With connectors, additional questions arise:

• Is the external source technically robust?
• Are Include and Exclude paths sensibly set?
• Is the maximum page number selected appropriately?
• Is the crawling frequency appropriate?
• Is re-crawling or re-indexing required after a change?

For files, the following points should be checked additionally:

• Is the file name unique?
• Is the file stored in the appropriate folder?
• Is the content current and technically checked?
• Has the file been reprocessed after changes?
• Is it clear whether the file is used in the original state or in a manually edited version?

Changes to the AI wiki not only affect the documentation. They can also affect the quality of response suggestions, chat contexts, and AI agents. Therefore, articles, connectors, files and visibility settings should be maintained as technical system components.