docs: add clear next-direction roadmap and improvement priorities (EN/ZH)

README.md

@@ -113,6 +113,24 @@ lite-sqlearner/
 └── package.json
 ```
 
+## 🧭 Next Direction
+
+The next phase of Lite-SQLearner focuses on making learning paths smarter, feedback clearer, and practice content richer:
+
+- **Adaptive learning path** — Recommend the next level based on pass/fail history and common mistakes
+- **Stronger answer diagnostics** — Explain *why* a query is wrong (syntax, logic, edge cases), not only pass/fail
+- **Content expansion** — Add intermediate/advanced topics (window functions, CTEs, query optimization, transactions)
+- **Practice quality tooling** — Add automated checks for level metadata, expected answers, and tutorial consistency
+- **Offline-first learning** — Improve local data persistence, progress export/import, and reliability without network
+
+## 🔧 Improvement Priorities
+
+If you want to contribute, these areas have the highest impact:
+
+1. **Learning experience**: richer hints, clearer error messaging, and progressive challenge design  
+2. **AI assistant quality**: prompt safety, model fallback behavior, and deterministic learning-focused responses  
+3. **Engineering quality**: add CI lint/test coverage, stabilize packaging workflow, and reduce regressions in releases
+
 ## 🤝 Contributing
 
 Contributions are welcome! Feel free to open issues or submit pull requests.

doc/README_zh.md

@@ -113,6 +113,24 @@ lite-sqlearner/
 └── package.json
 ```
 
+## 🧭 下一步方向
+
+Lite-SQLearner 的下一阶段将重点放在：学习路径更智能、反馈解释更清晰、练习内容更完整：
+
+- **自适应学习路径**：根据通关记录和常见错误推荐下一关
+- **更强的答案诊断**：不仅判断对错，还解释错误原因（语法、逻辑、边界情况）
+- **内容扩展**：补充中高级主题（窗口函数、CTE、查询优化、事务）
+- **练习质量工具化**：为关卡元数据、标准答案和教程一致性增加自动检查
+- **离线优先体验**：增强本地数据持久化、学习进度导入导出与无网可用性
+
+## 🔧 改进优先级
+
+如果你希望参与贡献，建议优先关注以下方向：
+
+1. **学习体验**：更好的提示机制、更清晰的报错说明、更平滑的难度曲线  
+2. **AI 助手质量**：提示词安全、模型降级策略、面向学习场景的稳定回复  
+3. **工程质量**：补齐 CI 的 lint/test 覆盖、稳定打包流程、降低版本回归风险
+
 ## 🤝 参与贡献
 
 欢迎提交 Issue 或 Pull Request！
@@ -122,4 +140,3 @@ lite-sqlearner/
 3. 提交你的更改（`git commit -m 'Add some amazing feature'`）
 4. 推送到分支（`git push origin feature/amazing-feature`）
 5. 发起 Pull Request
-