I benchmarked Claude Code and OpenCode on a heavy refactor: The reality of agentic CLI workflows
Every week, someone posts something like: “We are so cooked. AI has taken our jobs.” And every other week, someone else replies, maybe on X, saying that the same agent deleted their .env file, with the caption “lol.” Both reactions are understandable. Neither is enough evidence. To get a more useful answer, I benchmarked Claude Code…





