yashdeeptehlan

We don't track accept/reject yet but I just shipped preprompt-feedback in v0.1.2 that lets users rate optimizations.

It is not an extension. It learns your chat context and lives on your local machine.

Dialog clarification mode feature is going to be added in the next phase of development.

thank you for the suggestion.

Currently it dosen't track accept/reject yet but it's the right metric and i am adding it.

Thank you for your deep insight.

Thank you. i appreciate your compliment.

yes it is. it only requires your API credentials. everything stays local.

Hey HN, I'm Yashdeep — I built PrePrompt because I kept writing vague prompts and spending 3-4 turns correcting the model. I wanted something that fixed that automatically without changing my workflow. The classifier runs pure heuristics in <1ms — no API call. It only fires Haiku when a prompt scores above 38. Simple questions like "what is jwt" get a score of -5 and pass through untouched. The most interesting part: it learns your stack over time. After 50 prompts it knows you use FastAPI, prefer typed code, etc. and injects that into every optimization without being told. Happy to answer questions about the MCP hook architecture or the classifier scoring.