Files & Knowledge

Store files, organise them in folders, control visibility (private, company-wide, team, project) and use them as chat knowledge – with sources and page numbers.

In the Files area (sidebar) you store your documents centrally, organise them into folders and decide who may see them. Most importantly, they become the knowledge base for chat: Kasimir answers questions from them and backs every statement with source and page number. A team’s or project’s files are on the team or project page under Files & sources.

What this area does

• Upload and manage files – PDFs, Office and OpenDocument files, text, CSV, Markdown, HTML, XML, JSON, images and specialist formats such as GAEB, e-invoices, emails and CAD.

• Organise in folders – nested folders with an icon (emoji) and description.

• Control visibility – private, company-wide, team or project.

• Preview – PDFs, images and Office files right in the browser.

• Use as chat knowledge – Kasimir answers from them and cites the passages.

Uploading documents

1. Open the Files area from the sidebar.

2. Click Upload file at the top right and pick a file from your device.

3. The file uploads and is processed immediately (text is extracted). The status moves from Pending via Processing… to Ready – or to Failed if no text could be obtained. While a scan is going through text recognition, the file shows the progress instead (Reading text: page x of y).

Supported file types

Visibility on upload

There is no separate choice when uploading: if you upload into a folder, the file takes the folder’s visibility. Outside a folder the active tab decides – in the Shared tab the file becomes company-wide, otherwise private. Uploading company-wide requires the “Upload company knowledge” permission.

For team or project files, upload directly on the team or project page under Files & sources. A file’s visibility can be changed later (see below).

Folders & organization

Tabs, search and view

Above the file list you'll find:

• Tabs All · Shared · Private – filter folders and files by visibility.

• Search – filters the current level by file and folder names.

• View – toggle between Grid (with thumbnail) and List (with details). In both views every file has an actions menu.

Creating a folder

1. Click Folder.

2. Give it a name, optionally an icon (emoji) and a description.

3. Choose the visibility (Private or Shared).

4. When a folder is open, new subfolders and uploads automatically land inside it (shown in the breadcrumb at the top).

The three-dot menu of a folder lets you edit it (name, icon, description, visibility) or delete it.

Managing files

In both list and grid view, the three-dot menu per file offers these actions:

• Change visibility – switch between Private, Share company-wide and one of your teams. Who may do this is listed below under “Limits & good to know”.

• Move to – into another folder or back to General (no folder).

• Delete – with a confirmation prompt.

In grid view you can also drag files to move them: onto a folder or onto an entry of the breadcrumb at the top – that is also how you take a file back out of a folder.

Preview

Clicking a file (status Ready or Failed) opens the preview:

• PDFs and images are embedded and shown directly.

• Office / OpenDocument files are converted to a PDF preview server-side and shown the same way.

• All other formats offer a download of the original.

The preview also shows file size, type and a Download button for the original file.

Recently used (from chats)

At the very bottom – only at the root level and outside the Shared tab – you'll see Recently used (from chats). These are files you attached in chats yourself; the list only shows your own attachments. The attachment itself can be seen by anyone who may see the chat – in a project or team chat, that means all members. Per file you can:

• use Save to library to keep the file permanently (private) in your documents,

• or delete it permanently.

Using documents as chat knowledge

This is the core: your files become the AI’s knowledge source. There are two ways:

• Switch on company knowledge: in the bar below the input field, open Knowledge sources and turn on Company knowledge. Kasimir then searches all processed files visible to you. In a project or team chat, the files of the space are included automatically – you can switch that off for a single chat via the space button on the left of the bar; Company knowledge adds the company-wide files there.

• Attach a file: drag a file into the chat or attach it via +. It applies only to this chat; if all attachments fit the model, Kasimir reads them in full.

If a file is still being read (e.g. OCR of a scan), Kasimir waits with the answer and shows the progress; Answer now with what is available continues immediately, and the answer then states explicitly what is still missing.

Citations

When Kasimir uses individual passages from your documents, numbers like [1], [2] appear in the text. Below the answer, Sources opens the references with file name, page and excerpt – a click jumps to the cited passage.

Agents

In a chat with an agent, the agent’s knowledge (linked files and its own uploads) applies in addition. See Agents.

How Kasimir reads your documents

Kasimir decides by size: it reads small documents completely, and for large collections it picks the most relevant passages for each question. Under Model below the answer you can see which route was used (“Full context” or “Relevant passages”).

Search is tuned for German and also finds parts of compound words – “Sockel” also finds “Sockelhöhe”.

Setting the strategy per project

In project chats the strategy can be fine-tuned. Inside the project, the Settings tab has a Document context section with:

• Three tiles Decide automatically · Go deep · Find what's relevant (Hybrid and a custom threshold live under Advanced settings).

• A capacity meter that compares your documents' token usage against the active model's usable budget (including a rough page estimate, ~600 tokens/page).

• A breakdown of "in project" vs. "company-wide" plus the number of already indexed documents.

Visibility at a glance

Limits & good to know

• Maximum file size: 25 MB. Larger files are rejected.

• Pure images as company knowledge only work via the image description (text is derived from them); pixels without text yield little.

• Status Failed means no text could be obtained. Kasimir reads scanned PDFs, TIFF scans and images automatically via text recognition. A failed file can still be opened and downloaded but is not usable as a knowledge base.

• Changing visibility: for private files only the uploader may do this; for team files also the team's editors and leads; for company-wide files also owners and admins. “Share company-wide” requires the “Upload company knowledge” permission. Files from connected folders cannot be switched.

• Token values are estimates; the actual strategy may vary slightly by model.

• Search on the overview page filters file/folder names only (not full text) – content search happens in the chat via excerpt retrieval.

Which files Kasimir reads

Besides PDF, Office and images, since 16 Sept 2026 Kasimir also reads four groups of specialist formats: bills of quantities (GAEB), e-invoices (XRechnung and ZUGFeRD), e-mail files (.eml and .msg) and CAD drawings (.dxf and .dwg). Medical documents (HL7 FHIR and CDA) are possible but have to be released by an administrator.

These formats are not merely skimmed as text but read structurally — BoQ items, invoice lines, mail attachments, layers and dimensions of a drawing. That is why citations jump to the exact spot, and every format has its own preview.

The full list including limits and preview behaviour is at File formats in detail.