What problem does it solve?
It reduces uncertainty by quickly stress-testing an idea or plan, so you can reach the smallest useful next artifact without hidden assumptions or slow back-and-forth.
Core Features & Use Cases
- MVP-focused decision acceleration: Pushes toward the fastest valid path to an outcome (plan, decision, command, code, design, or recommendation).
- Mode-gated interaction: Selects target-file, in-memory, or patch-unit approaches depending on whether a user wants file edits immediately or prefers collecting changes before applying them.
- Uncertainty-first responses: Resolves factual uncertainty through reading, searching, fetching, inspecting, or testing before asking the user anything.
- Blocker-only questioning: Asks at most one blocker per turn and avoids permission-style or verification-choice question shapes.
Use it when you need to:
- stress-test direction for a feature, prototype, or design
- clarify assumptions and reduce scope
- challenge a plan without running a long interview
- decide the smallest coherent set of changes to produce next
Quick Start
Ask the AI to run an MVP grill for your feature idea: "Grill my MVP plan—stress-test assumptions, lock the fastest valid path, and ask only the one blocker you need to move forward."