I benchmarked Claude Code and OpenCode on a heavy refactor: The reality of agentic CLI workflows

In a recent benchmark comparing two agentic coding tools, Claude Code and OpenCode, both tools were tasked with refactoring a deeply nested Next.js 15 and React 19 dashboard from prop drilling to Zustand. The test aimed to evaluate how each tool managed the complex, multi-file migration involving TypeScript, shared UI state, and nested components. Both tools successfully completed the refactor and encountered the same environment issues, but OpenCode executed a cleaner first pass without introducing code-level mistakes, completing the task in 7 minutes compared to Claude Code's 14 minutes. The benchmark highlighted the importance of refactor order, explicit logging, and verification in using CLI coding agents, emphasizing that while these tools can accelerate development, they do not replace the need for engineering judgment. The key takeaway is that effective use of these tools requires developers to clearly specify migration strategies, identify potential risks, and ensure that a passing build maintains intended behavior.