> For the complete documentation index, see [llms.txt](https://operations.clausebase.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://operations.clausebase.com/updates-july-2024/14-july-2024.md).
# 14 July 2024
## Automatic Q\&A preparation
New feature In ClauseBuddy, administrators can automatically "prepare" the active Word-document for use as a Smart Template. For example, paragraphs such as the following...
... would be automatically replaced as follows:
This major new feature shows up as an extra element in the *Create new template* dropdown, assuming the end-user has the right to create new Smart Templates and has the right to use LLMs. (The LLM module that is used, is "Document analysis").
What happens under the hood, is that the LLM receives a copy of the active Word-document, and tries to detect relevant text sections that should be replaced by cyan-highlighted placeholders. These cyan placeholders can then, in a next phase, be used for creating questions in the Q\&A.
Note that this feature may also having interesting use-cases outside of Q\&A preparation, as the document is automatically anonymized in a fairly smart way.
{% hint style="info" %}
Do note that the anonymisation is not perfect, because this is quite a challenge for LLMs. In our tests with multiple LLMs, we notice that some glitches will inherently occur and that some cleanup will be necessary. Besides some reasoning errors that humans would not make, there are of course also some "gray areas" where humans would also disagree on whether a placeholder should actually be inserted.
{% endhint %}
## Default user preferences
New feature In both ClauseBuddy and Clause9, administrators now have a "New user defaults", which allows them to set the default settings for several individual user preferences (software language, content languages & currencies ordering, which elements must be inserted upon paste, whether personal search folders must be created, etc.)
Instead of merely setting the defaults for new users, administrators can also remove all current preferences of all existing users, by pressing the "Save for all users" button.
Administrators can also configure all those preferences per individual account.
***
## LLM-assisted Q\&A answers completion
New feature In both ClauseBuddy and Clause9, users can now paste text that will be used by an LLM to auto-fill the answers of the Q\&A.
For example, if a lawyer would get an email describing the type of license agreement that must be drafted, then she can copy/paste the relevant part of that email and automatically have (a subselection of the Q\&A's) questions be completed using that information
Note: the end-user must have the "Fill answers with AI" right enabled (not enabled by default), otherwise this button will not show up. The relevant LLM module is "clause analysis".
***
## Upgrade notice
Before the ClauseBase platform is updated (typically on a Saturday or Sunday morning in the CE(S)T timezone), a warning will be displayed that a server update is imminent.
After the update (which typically takes about one minute) users will see a banner inviting them to reload Clause9 or ClauseBuddy.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## 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, and the optional `goal` query parameter:
```
GET https://operations.clausebase.com/updates-july-2024/14-july-2024.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
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.