This article should answer:

Playbooks let you review legal documents using AI combined with your organisation’s own review rules.

Instead of relying only on general AI knowledge, Playbooks apply your internal rules to a document. They analyse clauses and paragraphs, check them against your requirements, and return structured feedback with references to the relevant sections.

The result: faster reviews that reflect how your legal team actually works.

Large Language Models (LLMs) like GPT are trained on publicly available information. While they understand legal language well, they do not know:

Playbooks solve this by defining your rules and (optionally) asking deal-specific questions before the review. This ensures the AI applies your standards - not generic assumptions.

For manual, non-AI reviews, use Checklists instead.

You can create Playbooks in several ways:

Create a blank category and add rules manually.

Use AI to turn existing text (e.g. internal playbooks or client guidelines) into rules. Simply select text in a document and let LawVu Draft generate a first version.

LawVu Draft provides ready-made playbook samples (Word documents). These are a practical starting point, especially if your team doesn’t already have structured rules.

These templates are intentionally imperfect - they may include too many or even conflicting rules. The goal is to review them internally, remove what’s not relevant, and refine what remains.

This method does not use GenAI, so formatting must be followed strictly.

For best results, format your Word document as follows:

If your document wasn’t based on a LawVu Draft sample (see previous option), you can still import it if it uses proper outline levels (e.g. Heading styles). Higher levels become groups; the lowest level becomes individual requirements. To preserve ordering, consider adding numbering (e.g. 01, 02, 03).

Within a category, you can add different types of elements:

A requirement defines what should be checked in a document.

You can also add extra elements via the + Augment button:

As described above, you can use the +Augment button to add various elements to your Requirement. Below, we describe these in more detail:

Internal guidance and support for colleagues, including relevant links, key contacts, and practical tips for carrying out tasks effectively. This information is intended strictly for internal use and is not shared with or processed by the LLM.

Specify the conditions under which the requirement applies. Multiple conditions can be combined, allowing you to define whether all, any, or none of them must be met for the requirement to take effect.

Clearly define the actions or outcomes that should follow the review process, specifying what needs to happen once the reviewing has been completed.

Allows you to link a clause from the Quality Library to the requirement. After reviewing, you can easily insert this clause into the active document or use it to compare (and even perform a smart merge) with the document’s current clause.

Allows you to rewrite the selected text from the active document after reviewing it, based on a predefined prompt or instruction.

It is important to note that this Action is not required to have LawVu Draft rewrite text based on the r-Requirement. This Action is intended to provide additional instructions to the LLM.

In other words, if the Requirement itself contains enough information to rewrite the relevant clause, this Action is not necessary.

Insert a comment into the document using predefined content, ensuring consistency and standardization in the information that is added.

Gathers and captures user input to support the evaluation of rules, ensuring that the necessary information is available to assess whether specific conditions or criteria are met.

Lets you add alternative positions to a requirement.

You can define layers such as Mostly met, Partially met, Not met, or Irrelevant to capture flexibility.

Multiple layers can be applied to a single requirement, reflecting different acceptable outcomes.

The Irrelevant layer marks a requirement as not applicable in specific situations.

During review, all layers are considered, showing which aspects are fully, mostly, partially, or not met.

This allows for nuanced evaluation rather than a simple yes/no outcome.

It helps capture internal variations and ensures that alternative positions are properly accounted for.

This feature checks for exact text, using traditional (non-LLM) matching.

Use it when wording must be precise (e.g. mandatory legal clauses). Keep in mind:

For multiple clauses, create separate rules rather than one long match.

Controls when a requirement should apply - or be ignored - by the LLM.

You could describe conditions in the Requirement itself, but using a separate Condition is recommended. It keeps things clearer and lets you add questions for the user (e.g. deal size), so the LLM can take deal-specific context into account.

When creating a condition:

Provides background information to help the LLM understand your content.

Use Insights to explain internal concepts or context that apply across multiple requirements, instead of repeating the same explanation each time.

An Insight includes:

Let you collect input from the user to help decide whether a requirement applies.

The answers are passed to the LLM, so it can take deal-specific information into account - especially when that information isn’t in the document (e.g. deal value).

You can add multiple questions, using different formats (e.g. yes/no, text, number, date, duration, currency).

Let you choose which rules apply to a specific type of deal.

Once your category contains multiple rules, you can group them into review sets.

Review sets allow you to:

This is useful when not all rules apply to every deal. For example, a low-value contract may require fewer checks than a high-risk, high-value agreement.

If no review set is used, all rules in the category are applied by default.

LLMs work best with clear, focused instructions. Avoid long or complex rules—think of the LLM like a junior colleague: clear key points are more effective than too much detail.

Focus on information the LLM may not know (e.g. internal policies, recent developments, niche areas). You can usually skip widely known or standard legal concepts unless precision is important.

AI-based review is still evolving. Playbooks should be used as a structured first-pass to help you:

Do not rely on the output alone—final legal judgement remains the responsibility of a qualified professional. Results will continue to improve as AI evolves.