Show HN: 规范驱动的 AI 开发 – 保持 AI 生成代码的可维护性
2 分•作者: samarthahathwar•6 个月前
我构建了一个系统来解决 AI 辅助编码的最大问题：六个月后你无法维护代码，因为你已经失去了关于代码编写原因的所有上下文。
核心思想很简单：不是直接跳入代码，而是先生成规范，然后用 Markdown 文件跟踪每一个规划决策和待办事项，这些文件在文件夹中移动（计划 → 进行中 → 已执行）。这个执行历史将永远伴随你的项目。
该工具包包括：
* 规范生成器（从需求生成架构和开发规范）
* 会话管理（跟踪跨编码会话的进度）
* 执行历史跟踪（3 文件夹系统）
* Git 工作流程助手（分支、规范提交）
* Spring Boot 测试生成
* 自动化代码审查代理（OWASP 安全、SOLID、性能）
它专为 Claude Code 构建，但工作流程概念适用于任何 AI 编码工具。
我用它构建了一个完整的 Spring Boot 后端——将 45 天的预估时间压缩到 5 天。几个月后，执行历史仍然可读。
完整工具包售价 49 美元：[https://hathwar.gumroad.com/l/spec-driven-ai](https://hathwar.gumroad.com/l/spec-driven-ai)
包含完整细节的登陆页面：[https://samhath03.github.io/spec-driven-ai/](https://samhath03.github.io/spec-driven-ai/)
欢迎就工作流程提出问题。
查看原文
I built a system to solve the biggest problem with AI-assisted coding: you can't maintain the code 6 months later because you've lost all context about WHY it was written.<p>The core idea is simple: instead of jumping straight into code, you generate specifications first, then track every planning decision and todo in markdown files that move through folders (plans → in-progress → executed). This execution history stays with your project forever.<p>The toolkit includes:
- Spec generators (architect + development specs from requirements)
- Session management (track progress across coding sessions)
- Execution history tracking (the 3-folder system)
- Git workflow helpers (branches, conventional commits)
- Spring Boot test generation
- Automated code review agents (OWASP security, SOLID, performance)<p>It's built for Claude Code but the workflow concepts apply to any AI coding tool.<p>I used this to build a complete Spring Boot backend — 45-day estimate compressed to 5 days. The execution history is still readable months later.<p>$49 for the complete toolkit: <a href="https://hathwar.gumroad.com/l/spec-driven-ai" rel="nofollow">https://hathwar.gumroad.com/l/spec-driven-ai</a><p>Landing page with full details: <a href="https://samhath03.github.io/spec-driven-ai/" rel="nofollow">https://samhath03.github.io/spec-driven-ai/</a><p>Happy to answer questions about the workflow.