docs: Refine security notice for YOLO mode usage

- Clarify YOLO mode should only be used in trusted environments on non-critical projects
- Add strong recommendation for --safe mode when working with important files
- Add guidance for using --safe mode with questionable model capabilities
- Update both English and Chinese README files

README.md

@@ -23,7 +23,7 @@ Use `# Your documentation request` to generate and maintain your AGENTS.md file
 
 Kode is a powerful AI assistant that lives in your terminal. It can understand your codebase, edit files, run commands, and handle entire workflows for you.
 
-> **⚠️ Security Notice**: Kode runs in YOLO mode by default (equivalent to Claude's `--dangerously-skip-permissions` flag), bypassing all permission checks for maximum productivity. This is recommended only for trusted environments with no internet access. For security-sensitive environments, use `kode --safe` to enable permission checks.
+> **⚠️ Security Notice**: Kode runs in YOLO mode by default (equivalent to Claude's `--dangerously-skip-permissions` flag), bypassing all permission checks for maximum productivity. YOLO mode is recommended only for trusted, secure environments when working on non-critical projects. If you're working with important files or using models of questionable capability, we strongly recommend using `kode --safe` to enable permission checks and manual approval for all operations.
 > 
 > **📊 Model Performance**: For optimal performance, we recommend using newer, more capable models designed for autonomous task completion. Avoid older Q&A-focused models like GPT-4o or Gemini 2.5 Pro, which are optimized for answering questions rather than sustained independent task execution. Choose models specifically trained for agentic workflows and extended reasoning capabilities.
 

README.zh-CN.md

@@ -7,7 +7,7 @@
 
 Kode 是一个强大的 AI 助手，运行在你的终端中。它能理解你的代码库、编辑文件、运行命令，并为你处理整个开发工作流。
 
-> **⚠️ 安全提示**：Kode 默认以 YOLO 模式运行（等同于 Claude 的 `--dangerously-skip-permissions` 标志），跳过所有权限检查以获得最大生产力。建议仅在无互联网访问的可信环境中使用。对于安全敏感环境，请使用 `kode --safe` 启用权限检查。
+> **⚠️ 安全提示**：Kode 默认以 YOLO 模式运行（等同于 Claude 的 `--dangerously-skip-permissions` 标志），跳过所有权限检查以获得最大生产力。YOLO 模式仅建议在安全可信的环境中处理非重要项目时使用。如果您正在处理重要文件或使用能力存疑的模型，我们强烈建议使用 `kode --safe` 启用权限检查和手动审批所有操作。
 > 
 > **📊 模型性能建议**：为获得最佳体验，建议使用专为自主任务完成设计的新一代强大模型。避免使用 GPT-4o、Gemini 2.5 Pro 等较老的问答型模型，它们主要针对回答问题进行优化，而非持续的独立任务执行。请选择专门训练用于智能体工作流和扩展推理能力的模型。