A chemistry robot, a Kepler agent and an alloy checker in one research direction

A chemistry robot, an agent that rediscovers Kepler's laws from raw data, and a system that works out whether an alloy can be 3D printed all landed in the same Lacuna research direction.
They all rely on the same methodology: an LLM takes a high-level scientific goal and translates it into work for a set of specialized tools.
Marta Skreta's CLAIRify turns a chemistry instruction into task-and-motion plans for a robot. PhysAgent coordinates agents to recover physical laws from data. The alloy system has Claude call Thermo-Calc and a melt-pool simulator to find safe combinations of laser power and scanning speed.
Lacuna found the connection by breaking each paper into concept elements, usually a sentence or two per idea, and clustering their embeddings. This particular research direction contains 50 concepts from 43 papers.
Later this week, we'll share a way to point your own agents at the same map.





First posted on LinkedIn, July 21, 2026.


