Building a Public Sector AI Tools Catalogue with Cursor

TL;DR: Check out this Public Sector Tools catalogue I made with Cursor.

In my latest role with the UK’s Government Digital Service, we’re running a “horizon scanning” function to understand what AI-enabled tools teams are developing across the UK public sector.

Prior to my joining, the team built an AI-powerd web-scraper using Python, with CrewAI running on AWS Bedrock. It periodically searches the web to produce an augmented, updated, and correct tools dataset and an HTMl front-end.

The front-end includes a fun visualisation using vis-network.js and html-to-image.js.

It looks like this.

Zoom in a bit and you can see the network graph edges.

You can click individual nodes and see information about them. It’s fun and looks cool, but it’s not particularly easy to search or browse.

Building a GOV.UK-style browsable site

One nice feature of the original GDS site is that it allows you to download a CSV file of all the data.

So I downloaded it, opened Cursor, and gave it a the following very sloppy prompt (I’ve kept the typos, GenAI doesn’t care):

" Eleventy (11ty) to create a static site usign the GOV.UK design system and styles. The site should allow users to browse a catalogue of AI tools in use within the public sector. It should enable them to browse by organisation, or AI method with a single page for each row in a supplied CSV file. Here is a snippet of the file showing the headers and a few rows of data. What questions do you have?"

I also included the header and about 5 rows of data from the CSV.

I was impressed by the insightful questions it raised. It even gave me an easy out by telling me which questions were most important to answer enabling me to provide a “good enough” response to get a decent first pass.

I encountered a couple of minor errors with dependency resolution and a deprecation warning, but with minimal back-and-forth, I had a working site. I then directed how I wanted the navigation and search to work, fixing a few small bugs. I was amazed at how quickly I built the site the site came together and how readable the code was. This is my first time using 11ty.js, and while I understand about 60% of what Cursor created, I was impressed with its explanations of design decisions and implementation details.

I deployed everything to AWS Amplify, and within minutes, I had a live site with a content delivery network and HTTPS connection that rebuilds every time I push to Github. I used Cursor’s built-in composer model, which is much cheaper than big ones like Opus.

I was really impressed by how easy this was, and now we have a much more navigable website. Visit the site or check out the code. It’s surprisingly elegant and I wrote almost none of it. I’d love to hear your thoughts!

A quick disclaimer: I’m using data from the UK Government Digital Service, and while I’m currently working with them, the catalogue website is made by me and not endorsed by GDS or the Department for Science, Innovation, and Technology in any way.