> 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-january-2025/6-january-2025.md).

# 6 January 2025

## Search result optimisations in Truffle Hunt

<mark style="background-color:orange;">New major feature</mark> You can now use GenAI to refine your search results. By clicking on *“Optimize search results”* button and then optionally adding an additional prompt, all your results will be sent to an LLM, who will then intelligently match your search query + prompt with your search results, and have the results automatically be reordered for relevance.&#x20;

<figure><img src="/files/ttWhG5GyvN2In98apgd3" alt="" width="563"><figcaption></figcaption></figure>

Moreover, the LLM will add interesting highlights in the search results, to allow you to quickly see why certain results are indeed relevant.

You can particularly use this as an intelligent additional filter. For example, if you’re looking for a specific type of content, you can submit this as your additional prompt.&#x20;

Going forward, the best strategy is to do a rough query first and then refine your results, either by comparing them with “Extract legal features”, or by using the new “Optimize search results”. (If you’re wondering: the reason for this two-step approach is that LLMs are simply too slow to do the initial selection, and they are not able to handle thousands of clauses. You must therefore feed them a more limited set.)

## Search result optimisations in AutoSuggest

<mark style="background-color:orange;">New major feature</mark> The “Extract legal features” option is now also possible in the keyword-based search mode of the Quality Library. (Previously, for technical reasons, it was only available in the Browse-mode. (You’ll also notice that we slightly rearranged the icons here.)

<figure><img src="/files/gum3IvHISVkuITf4tZZh" alt="" width="563"><figcaption></figcaption></figure>

***

## Better DOCX layout insertions

<mark style="background-color:orange;">New major feature</mark> While competing software packages simply copy/insert plain text, ClauseBuddy always tried to go the extra mile, by doing a “guesstimate” of the documents’ layout, and/or offering the possibility to customers to define their own custom house-styling. This worked reasonably fine for documents with good layout... which unfortunately only represent perhaps 30% of all legal documents. &#x20;

After months of development, we now introduce **a new layout engine that performs a very deep analysis of the active Word-document**.&#x20;

The results are much better, particularly for documents with poor layout. Furthermore, everything is faster and works without any server/LLM involvement, which is important from a security perspective.&#x20;

Some examples:

Titles that split in the middle are usually correctly used:

<figure><img src="/files/eJ86URDGN0oe5u60q0Q3" alt="" width="563"><figcaption></figcaption></figure>

“Embedded titles” (e.g. definitions with the term in bold) are usually recognised as such:

<figure><img src="/files/6qurhHcOLSFBxACBR3Fp" alt="" width="563"><figcaption></figcaption></figure>

ClauseBuddy does not fix bad layouts, and has to do a lot of guessing, so this simply cannot be perfect and will still go wrong often. But usually the results are usually at least “OK”.

A few additional changes to mention:

* To reduce clutter, the *“Insert with changes”* option only appears when some text is selected in MS Word
* There’s always a Copy button (to copy in plaintext), but be aware that it doesn’t work for Truffles on Mac (error message).
* When inserting paragraphs between manually numbered paragraphs, the number deliberately shows up in red to highlight that

<figure><img src="/files/lGVBj1YrAKFJFjkYFh1R" alt="" width="207"><figcaption></figcaption></figure>

* The icons at the bottom are simplified; the few relevant options are accessed through a submenu.&#x20;

<figure><img src="/files/2iv0fME1h98jwMyktBLX" alt="" width="563"><figcaption></figcaption></figure>


---

# 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-january-2025/6-january-2025.md?ask=<question>&goal=<endgoal>
```

`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.