Integrating LAWMA with WordPress requires more than adding a single script. In one implementation, the complete setup included loading the LAWMA Thin Client site-wide, placing the FAQ Agent only where it was needed, and publishing two supporting files at the web root.
The final configuration kept the AI-powered FAQ widget on the dedicated FAQ page rather than in the global header, which helped avoid unnecessary DOM growth and page-speed impact. The required llms.txt and sdl.json files were also uploaded and verified as publicly accessible.
Issue background
A WordPress site received a LAWMA implementation guide and needed to confirm exactly what changes were required on the website.
The documented implementation called for four core pieces:
- The LAWMA Thin Client JavaScript snippet loaded in the site-wide
<head>. - The LAWMA FAQ Agent container placed on any page where the AI-powered FAQ should appear.
- An
llms.txtfile uploaded to the website root. - An
sdl.jsonfile uploaded to the website root.
The team also needed to decide where the FAQ widget should live and whether it should be surfaced globally near the site’s search interface.
Diagnosis
The developer reviewed the LAWMA implementation guide and confirmed that those four items covered the complete technical website implementation.
The FAQ widget was not designed to appear automatically across the site. It renders only where the following container is added:
<div id="display-lawmafaqai"></div>LAWMA also supports optional FAQ category filtering through a data-label attribute, for example:
<div id="display-lawmafaqai" data-label="services"></div>The team considered placing the FAQ widget in or near the global header, but the developer advised against that approach. Loading the full FAQ interface globally would increase the DOM size on every page and could negatively affect page-speed scores.
The dedicated FAQ page was therefore the preferred location. Additional instances could be added later to relevant content pages as the site’s FAQ strategy expanded.
Resolution steps
The documented implementation process was:
- Add the LAWMA Thin Client once. Install the provided LAWMA JavaScript in the site-wide
<head>using the client API key supplied by LAWMA. Avoid loading the Thin Client more than once so the integration does not initialize twice. - Choose the FAQ Agent location deliberately. Add
<div id="display-lawmafaqai"></div>only on pages where the LAWMA FAQ interface should render. - Use the dedicated FAQ page as the initial placement. In this implementation, the FAQ page was selected as the best balance between visibility and performance.
- Use category filtering only if needed. If a page should display a narrower FAQ set, add the appropriate
data-labelvalue supported by LAWMA. - Upload
llms.txtto the web root. The file should be accessible directly from the site’s root path, such as/llms.txt. - Upload
sdl.jsonto the web root. This file should likewise be available directly from the root, such as/sdl.json. - Clear site and CDN caches. After deployment, purge caching so the new script, widget markup, and root-level files are served correctly.
- Verify the widget and both files. Confirm that the FAQ Agent renders on the intended page and that both root files are publicly accessible.
- Check the browser console. Review for JavaScript errors or Content Security Policy issues that could block the Thin Client or widget.
- Maintain the root files over time. LAWMA periodically regenerates
llms.txtandsdl.jsonas its content changes. Those updated copies are not automatically synchronized by the website snippet, so they need to be downloaded from the LAWMA portal and uploaded again when revised.
The implementation was completed with the FAQ Agent on the dedicated FAQ page, and the two supporting files were verified as accessible from the root of the site.
Final outcome
The LAWMA integration was completed successfully. The Thin Client was installed site-wide, the FAQ Agent displayed correctly on the intended page, and both llms.txt and sdl.json were accessible from the website root.
The team also established a sensible placement strategy: keep the FAQ widget on dedicated or contextually relevant pages rather than loading it globally in the header, where it could add unnecessary DOM weight to every page.
The broader lesson is that third-party AI integrations should be implemented with both functionality and performance in mind. Load global scripts only once, render heavier widgets only where they add value, and treat supporting root files as maintained assets rather than one-time uploads.
If you need help implementing LAWMA or another AI-powered FAQ integration on WordPress, contact Freshy. Our WordPress team can help place the scripts correctly, manage root-level files, and test the integration without adding unnecessary frontend overhead.
