exploratory-sota-research

What problem does it solve?

It maps the current AI/ML state of the art for a concrete technical problem into a mechanism-and-bottleneck landscape, so you can choose research directions based on evidence rather than a flat reading list.

Core Features & Use Cases

• Exploratory mechanism landscape: Decomposes the real decision into problem slices, hypothesized root causes, and mechanism families to search for generalizable approaches.
• Evidence-driven synthesis: Clusters findings by bottleneck and mechanism, actively seeks disconfirming evidence, and distinguishes reusable mechanisms from narrow hacks.
• Actionable next moves: Produces an explicit portfolio of recommendations (Now / Next / Explore / Avoid) with validation experiments and what could invalidate each direction.

Quick Start

Use the exploratory-sota-research skill to map the state of the art for: "Improve [your task] with measurable success criteria [metric] under constraints [latency/cost/data regime], and recommend which mechanism families are worth testing next."