prototype

You can validate product ideas by building a disposable prototype that tests specific technical uncertainties through isolated experiments. This approach requires defining a single named question and a one-command execution path to empirically verify UI interactions or state transitions before full-scale implementation.

A spike is a minimal, disposable experiment built to test specific product or technical uncertainties through isolated artifacts. You need one when you require empirical evidence for UI interactions or state transitions before committing to full-scale implementation.

You can test if a new interface layout confuses users by creating a single-screen disposable mockup. This prototype isolates the specific interaction in question, allowing you to gather empirical evidence and make evidence-based decisions before building the full feature.

Yes, you can run prototype experiments without external dependencies. The skill operates independently to build minimal, runnable artifacts designed to answer a single specific question and verify product assumptions through a one-command execution path.

The best way to prevent building full features on unverified assumptions is to build disposable experiments that force a clear definition of success and failure. This ensures prototypes are used as empirical evidence rather than production code, saving development time and reducing wasted effort.

You should not use a disposable prototype when you need production-ready code, as these artifacts are designed to be non-production and temporary. Avoid this approach if your task lacks a specific, single question or a defined observer to verify the empirical results.