Technology Radar
Anthropic's Claude Code is an agentic AI coding tool for planning and executing complex multi-step workflows. We’re moving Claude Code to Adopt because teams inside and outside Thoughtworks now use it day-to-day in production software delivery, where it's widely treated as a benchmark for capability and usability. The CLI agent landscape has expanded quickly with tools such as OpenAI's Codex CLI, Google's Gemini CLI, OpenCode and pi, but Claude Code remains the preferred option for many teams. Its use now extends beyond code authoring into broader workflow execution, including specifications, stories, configuration, infrastructure, documentation and markdown-defined business processes. Claude Code continues to introduce features that other tools follow, such as skills, subagents, remote control and agentic team workflows.
Teams adopting Claude Code should pair it with disciplined operating practices. Agentic coding shifts developer effort from manual implementation toward specifying intent, constraints and review boundaries. This can accelerate delivery, but it also increases the risk of complacency with AI-generated code, which can make systems harder to maintain and evolve for both people and agents. We're seeing growing attention to harness engineering as a way to implement context engineering (topic-aware, scope-driven context selection) and curated shared instructions to make agentic workflows more reliable.
Anthropic 的 Claude Code 是一款具备智能体特性的 AI 编码工具，它为规划和执行复杂的多步骤工作流提供了自然语言接口和智能体执行模型。自发布不到一年以来，它已经在 Thoughtworks 内部和外部的开发者中被广泛采用，因此我们将其纳入 Trial 阶段。基于命令行的编码智能体，如 OpenAI 的 Codex CLI、Google 的 Gemini CLI 以及开源项目 OpenCode 相继推出，而基于 IDE 的助手，如 Cursor、Windsurf 和 GitHub Copilot 也纷纷加入了智能体模式。即便如此，Claude Code 仍然深受欢迎。我们观察到，团队不仅使用它来编写和修改代码，还将其作为通用型 AI 智能体，用于管理规格说明、用户故事、配置、基础设施以及文档。 智能体式编码将开发者的关注点从“编写代码”转移到“表达意图并委派实现”。这虽然能够加快开发周期，但也可能导致对 AI 生成代码的自满，从而产生人类和 AI 智能体都更难维护和演化的代码。因此，团队必须严格管理 Claude Code 的使用方式，可采用诸如上下文工程、精选共享指令以及可能的编码智能体团队等技术手段来实现。
