Brief:
The brief today was to build an Alteryx app that builds an HTML trend analysis report based on a user input. We had the choice of two API's and I decided to go with the Wikimedia API which contains Wikipedia page info.
Process:
- Reading through API documentation and identifying specific format requirements. For example dates had to be formatted: 27/08/2026 ->27082026.
- Built Alteryx Workflow.
- Added user actions to turn into an app.
- Numerous prompt iterations.
Challenges:
- Main challenge was over engineering anytime I tried to be too specific and include every instruction for how I wanted things done the html output would get truncated and refuse to load. This was solved by simplifying my prompt.
- Date axis on charts not working had to manually force a library inclusion.
- LLM generation streams appended markdown code fences (````html`) to the top of the raw text response, rendering visible backtick artifacts on the published page. So I implemented string stripping rules in both the prompt instructions and an Alteryx
REGEX_Replace()clean-up formula prior to writing the.htmlfile.
Here is my final workflow:

And here is the user interface:

Overall this was my final output:
