Hello All,
I would like to share my app that I have built to test and evaluate prompts (LLM requests). It is a single .html file, BYOK (Bring Your Own Key) app that supports OpenAI and Gemini protocols. The API Keys are stored in your browser and never leave your PC; the code is open, so you can check it yourself.
You may think of it as the Postman for LLMs.
Specify the raw LLM request (JSON), write your prompt, define your test cases, and click Test!
I use it constantly to troubleshoot LLM requests and to make sure that my prompts consistently work as expected under different use cases.
Here is the link to the app, i hope it will be helpful to you too.
dmytronius.github .io/ prompt-tester/PromptTester.html
BR,
Dmytronius