Love Claude Code. Dislike, to the point of some resentment, that the code (detailing the harness, the system prompt, the agentic patterns) is not open-source and inspectable.
Have any opinions on Claude Code vs OpenCode or one of the other similar opensource tools? There are almost too many options now to evaluate.
There are proxies that let you inspect everything. Also some people managed to decompile it.
This is nothing short of brilliant. New styles can encourage to continue learning, while using AI.
I was sort of using this ad hoc before: asking Claude implementation questions after the coding was done. But these styles can take it to next level.