MLR Bot is designed to help content owners and coordinators prepare their documents for MLR review by detecting common issues before submission, such as spelling errors, prohibited phrases, and missing warnings or safety information. Leveraging a Large Language Model (LLM), MLR Bot proactively identifies a wide range of issues in your promotional materials, reducing the number of review cycles and accelerating time to approval.
MLR Bot is integrated directly within PromoMats, which means your documents and data do not leave your Veeva Vault environment, ensuring data stays secure, and user security is respected.
Note: This feature is currently available only to early access customers.
Accessing the MLR Bot Panel
Once configured by an Admin, the MLR Bot panel is available when all of the following conditions are met:
- The document type belongs to the MLR Bot document type group.
- The current document version is the latest version.
- The document does not exceed 20 pages in length.
- The document has a PDF viewable rendition and does not contain audio or video content.
- The document is not in the Steady, Superseded, or Obsolete states.
- The document is not archived.
- The document is not a binder or CrossLink.
While viewing a document in the document viewer, click 
to open the MLR Bot panel. You can click and drag to expand the width of the panel.
Note: The MLR Bot quality check outcomes are not audited, and are not reportable. Quality check results and Assistant conversations are private to the user and specific to the document version being assessed.
Performing Quality Checks
By default, the MLR Bot panel opens on the Quality Checks tab, and immediately begins performing the following predefined quality checks on the document:
- Spelling
- This check identifies potential spelling errors in the document, focusing on misspellings, common misuses, and typos, while considering industry-specific terminology.
- Grammar
- This check identifies potential grammatical errors in the document, such as verb tense, punctuation, pronoun and subject-verb agreement, sentence structure, and modifier placement.
- Red Flag Phrases
- This check identifies phrases that could indicate non-compliance with regulators, such as unsubstantiated claims, risk minimization, and off-label uses.
- Boxed Warning
- This check determines whether the document’s Product requires a Boxed Warning according to the FDA database of Boxed Warning products.
- If MLR Bot determines that a Boxed Warning is required, it evaluates the Warning’s presence, location, and general format.
- MLR Bot does not review the text of the Boxed Warning.
- Important Safety Information
- This check determines whether a related ISI document exists and, if so, whether the Important Safety Information in that document is reflected in the current document.
- MLR Bot assess ISI wording to identify omissions, additions, inaccuracies, or misrepresentations of critical safety information.
- If there is no related ISI document, MLR Bot determines this check to be Not Applicable.
- Privacy Policy Link
- This check determines whether the document requires a link to a privacy policy based on its content and intended use, and extracts links to determine if any privacy-related links are present and match existing Website object records.
- Unsubscribe Link
- This check determines whether the document requires an unsubscribe link, and extracts links to determine if any unsubscribe links are present and match existing Website object records.
- Accessibility
- This check validates whether the document conforms to web accessibility standards, identifying potential barriers for users with disabilities.
- MLR Bot evaluates structure and organization, language and readability, and assesses visual elements for text contrast and alt text.
Note: Veeva defines these quality checks. You cannot edit these checks or add custom ones.
Viewing Quality Check Results
When the quality checks are complete, MLR Bot displays an icon to indicate whether the check passed () or identified issues (). Click the check title to expand or collapse the section. MLR Bot displays if it encounters an error, such as when the document exceeds the page limit.
To navigate to an identified issue, click the page number to jump to the issue or the page the issue is on. By default, Vault does not save any detected issues to the document, but you can choose to save specific issues as comment annotations by clicking the Save As Annotation () button.
Matching Websites
When performing the Privacy Policy Link and Unsubscribe Link quality checks, MLR Bot attempts to match any extracted links to existing Website object records based on the Country and Language fields on the document and object records. If the Country and Language fields are blank on a Website record, MLR Bot considers the record to apply to all values.
Using the MLR Bot Assistant
MLR Bot also provides access to an intuitive AI Assistant, which you can access by clicking the Assistant tab in the MLR Bot panel. You can ask the Assistant questions about the specific document version you are viewing. MLR Bot generates responses based on the document’s text, images, and context from any completed quality checks.
To use the Assistant, enter your query in the text box at the bottom of the panel. Press Shift + Enter on your keyboard for a new line. and press Send 
or press or the Enter key. You can enter a maximum of 5,000 characters per query. MLR Bot displays 
while the Assistant generates a response.
Related Permissions
The following permissions control your ability to use MLR Bot:
| Type | Permission Label | Controls |
|---|---|---|
| Document Role | View Content | Ability to view and use the MLR Bot panel on a document version in a given lifecycle state |
| Document Role | Annotate | Ability to save an identified issue as a document annotation |