Created small web tool for generating syntetic test data for LLM prompts (Begginer project)

Hey everyone,

I recently started learning web development and put together a small single-page tool called Synthetic Context Engine.

The idea behind it is to quickly generate fake background documents, server logs, corporate emails, or gaming rules to test how well AI models find specific information in long prompts (needle in a haystack style).

Choose different document scenarios

Adjust text length and noise level

Hide a specific secret key or code at a chosen depth

Generate data locally or connect your Gemini API key directly

Since I am a beginner, I am pretty sure some things might be buggy or not working as intended yet because its starter version 0

I would really appreciate any feedback, suggestions, or advice on how to improve the code.

Link: https://arhistrategstudio.github.io/Syntetic-prompt-generator/