Kill My Idea
A web app asks Jev a set of parallel questions to decide whether a startup idea should be killed, fixed, or shipped.
A web app asks Jev a set of parallel questions to decide whether a startup idea should be killed, fixed, or shipped.
The server sends the startup idea as `state.startup_idea` with eight goal-adjusted Score questions, a Choice for product category, and a Noul for understandability. Code normalizes and weights the eight scores, applies a clarity gate, and maps the result to KILL, FIX, or SHIP.
Does `startup_idea` solve a real problem that people actually feel?
The request state has one field, `startup_idea`, containing the submitted idea; no example runtime value is asserted.
The server validates the idea text and selected goal.
Jev receives the idea and asks eight goal-adjusted Score questions alongside category and clarity questions.
Code normalizes the scores to 0–100, applies goal-specific weights and a clarity gate, then maps the total to KILL, FIX, or SHIP.
The app presents dimension results and a collapsible panel with raw Jev answers and evaluation metadata; successful evaluations can be archived.
The verdict is assembled from inspectable dimensions, goal-specific weights, and a separate clarity gate rather than a single free-form model opinion.