Use agent to learn agent - A skeleton course on how to design, build, and operate production AI agents
Terminal score: 0–1000 raw, weighted across 4 dimensions. Public score: 0–10 normalized (shown in the 30-day stars chart above).