leniu-dev 2.0.0

package/.claude/skills/collaborating-with-codex/SKILL.md

@@ -0,0 +1,180 @@
+---
+name: collaborating-with-codex
+description: |
+  与 Codex MCP 协同开发。通过 GuDaStudio/codexmcp 集成，使用 mcp__codex__codex 工具。
+  默认沙箱：read-only（严禁 codex 修改真实代码）。
+
+  触发场景：
+  - 需要算法实现或复杂逻辑分析
+  - 需要代码审查和 Bug 分析
+  - 需要生成 Unified Diff 补丁
+  - 用户明确要求使用 Codex 协作
+  - 复杂后端逻辑的原型设计
+
+  触发词：Codex、协作、多模型、原型、Diff、算法分析、代码审查、codex协同
+
+  前置要求：
+  - 已通过 `claude mcp add codex -s user --transport stdio -- uvx --from git+https://github.com/GuDaStudio/codexmcp.git codexmcp` 注册
+  - ~/.claude/settings.json 的 permissions.allow 已加入 mcp__codex__codex
+---
+
+# 与 Codex MCP 协同开发
+
+> 通过 `mcp__codex__codex` 工具直接调用，无需命令行。始终使用 `read-only` 沙箱，要求 codex 只输出 unified diff patch。
+
+---
+
+## MCP 工具说明
+
+### 工具名
+`mcp__codex__codex`
+
+### 参数
+
+| 参数 | 类型 | 必填 | 默认值 | 说明 |
+|------|------|------|--------|------|
+| `PROMPT` | string | ✅ | - | 发送给 Codex 的任务指令（建议用英语） |
+| `cd` | Path | ✅ | - | 工作目录根路径（必须存在） |
+| `sandbox` | string | ❌ | `read-only` | `read-only` / `workspace-write` / `danger-full-access` |
+| `SESSION_ID` | UUID | ❌ | None | 继续之前的会话，保持上下文 |
+| `return_all_messages` | bool | ❌ | False | 是否返回推理过程和工具调用详情 |
+| `model` | string | ❌ | None | 指定模型（None 使用用户默认配置） |
+| `image` | List[Path] | ❌ | None | 附加图片文件 |
+| `yolo` | bool | ❌ | False | 跳过沙箱审批（危险，不推荐） |
+| `profile` | string | ❌ | None | 从 `~/.codex/config.toml` 加载的配置名 |
+| `skip_git_repo_check` | bool | ❌ | False | 允许在非 Git 仓库中运行 |
+
+### 返回值
+
+```json
+// 成功
+{
+  "success": true,
+  "SESSION_ID": "uuid-string",     // 必须保存，用于多轮交互
+  "agent_messages": "codex回复内容"
+}
+
+// 失败
+{
+  "success": false,
+  "error": "错误信息"
+}
+```
+
+---
+
+## 标准协作流程
+
+```
+1. 需求分析阶段
+   Claude 形成初步分析 → 告知 codex → codex 完善方案
+
+2. 编码前（原型阶段）
+   调用 codex（read-only）→ 要求输出 unified diff patch
+   Claude 以 diff 为逻辑参考 → 重写为生产级代码
+
+3. 编码后（审查阶段）
+   调用 codex review 改动 → 验证需求完成度和代码质量
+```
+
+---
+
+## 调用示例
+
+### 场景一：需求分析完善
+
+```
+PROMPT: "Here is my requirement: [需求描述]. My initial approach: [初步思路].
+Please review and improve the analysis and implementation plan.
+Output: structured analysis with potential risks."
+
+cd: /path/to/project
+sandbox: read-only
+```
+
+### 场景二：获取代码原型（Diff）
+
+```
+PROMPT: "Generate a unified diff patch to implement [功能描述].
+File: [目标文件路径]
+Requirements:
+- [要求1]
+- [要求2]
+
+IMPORTANT: Output unified diff patch ONLY. Do NOT modify any real files.
+IMPORTANT: All Java comments and SQL COMMENTs MUST be in Chinese."
+
+cd: /path/to/project
+sandbox: read-only
+```
+
+### 场景三：代码审查
+
+```
+PROMPT: "Review the following code changes for correctness, performance, and security.
+[粘贴代码或说明文件路径]
+
+Check:
+1. Logic correctness
+2. Edge cases
+3. Performance issues
+4. Security vulnerabilities
+
+Output: review with line numbers and severity (CRITICAL/WARNING/INFO)."
+
+cd: /path/to/project
+sandbox: read-only
+```
+
+### 场景四：多轮交互（继续会话）
+
+```
+// 第一轮
+SESSION_ID: None → 保存返回的 SESSION_ID
+
+// 第二轮
+SESSION_ID: "上一轮返回的 uuid"
+PROMPT: "Now refine the diff based on the feedback: [反馈内容]"
+```
+
+---
+
+## 协作分工原则
+
+| 角色 | Claude Code 负责 | Codex 负责 |
+|------|-----------------|-----------|
+| **架构** | 设计决策、规范审校 | 分析现有代码结构 |
+| **开发** | 规范重写、最终代码实施 | 输出原型 diff（只读参考） |
+| **审查** | 规范检查、最终判定 | 逐文件逻辑审查 |
+| **调试** | 日志分析、问题定位 | 深度代码分析、补丁建议 |
+
+### 重要约束
+
+1. **只读优先**：始终使用 `sandbox="read-only"`
+2. **英语 Prompt**：与 codex 交互用英语，代码注释要求中文
+3. **脏原型思维**：codex 输出视为草稿，Claude 按项目规范重构
+4. **保存 SESSION_ID**：每次调用后记录，多轮对话时传入
+5. **质疑 codex**：codex 仅供参考，必须有独立判断
+
+---
+
+## 故障排除
+
+| 问题 | 解决方案 |
+|------|---------|
+| 工具未出现 | 重启 Claude Code，检查 `~/.claude.json` 中 codex 配置 |
+| `cd` 参数报错 | 确认目录存在，使用绝对路径 |
+| 连接超时 | 检查网络，`uvx --from git+https://github.com/GuDaStudio/codexmcp.git codexmcp` 手动测试 |
+| SESSION_ID 失效 | 重新开启新会话（不传 SESSION_ID） |
+| codex 修改了文件 | 确认 `sandbox="read-only"` 已设置 |
+
+## 安装验证
+
+```bash
+# 查看 MCP 配置
+cat ~/.claude.json | python3 -c "import json,sys; d=json.load(sys.stdin); print(d.get('mcpServers', {}).get('codex', 'not found'))"
+
+# 重启后运行
+claude mcp list
+# 期望看到: codex: uvx ... - ✓ Connected
+```

package/.claude/skills/collaborating-with-codex/scripts/codex_bridge.py

@@ -0,0 +1,275 @@
+"""
+Codex Bridge Script for Claude Agent Skills.
+Wraps the Codex CLI to provide a JSON-based interface for Claude.
+"""
+from __future__ import annotations
+
+import json
+import re
+import os
+import sys
+import queue
+import subprocess
+import threading
+import time
+import shutil
+import argparse
+from pathlib import Path
+from typing import Generator, List, Optional
+
+
+def _get_windows_npm_paths() -> List[Path]:
+    """Return candidate directories for npm global installs on Windows."""
+    if os.name != "nt":
+        return []
+    paths: List[Path] = []
+    env = os.environ
+    if prefix := env.get("NPM_CONFIG_PREFIX") or env.get("npm_config_prefix"):
+        paths.append(Path(prefix))
+    if appdata := env.get("APPDATA"):
+        paths.append(Path(appdata) / "npm")
+    if localappdata := env.get("LOCALAPPDATA"):
+        paths.append(Path(localappdata) / "npm")
+    if programfiles := env.get("ProgramFiles"):
+        paths.append(Path(programfiles) / "nodejs")
+    return paths
+
+
+def _augment_path_env(env: dict) -> None:
+    """Prepend npm global directories to PATH if missing."""
+    if os.name != "nt":
+        return
+    path_key = next((k for k in env if k.upper() == "PATH"), "PATH")
+    path_entries = [p for p in env.get(path_key, "").split(os.pathsep) if p]
+    lower_set = {p.lower() for p in path_entries}
+    for candidate in _get_windows_npm_paths():
+        if candidate.is_dir() and str(candidate).lower() not in lower_set:
+            path_entries.insert(0, str(candidate))
+            lower_set.add(str(candidate).lower())
+    env[path_key] = os.pathsep.join(path_entries)
+
+
+def _resolve_executable(name: str, env: dict) -> str:
+    """Resolve executable path, checking npm directories for .cmd/.bat on Windows."""
+    if os.path.isabs(name) or os.sep in name or (os.altsep and os.altsep in name):
+        return name
+    path_key = next((k for k in env if k.upper() == "PATH"), "PATH")
+    if resolved := shutil.which(name, path=env.get(path_key)):
+        return resolved
+    if os.name == "nt":
+        for base in _get_windows_npm_paths():
+            for ext in (".cmd", ".bat", ".exe"):
+                candidate = base / f"{name}{ext}"
+                if candidate.is_file():
+                    return str(candidate)
+    return name
+
+
+def run_shell_command(cmd: List[str], stdin_data: Optional[str] = None) -> Generator[str, None, None]:
+    """Execute a command and stream its output line-by-line.
+
+    Args:
+        cmd: Command and arguments as a list
+        stdin_data: Optional string to pass to the process via stdin
+    """
+    env = os.environ.copy()
+    _augment_path_env(env)
+
+    popen_cmd = cmd.copy()
+    exe_path = _resolve_executable(cmd[0], env)
+    popen_cmd[0] = exe_path
+
+    process = subprocess.Popen(
+        popen_cmd,
+        shell=False,
+        stdin=subprocess.PIPE if stdin_data else subprocess.DEVNULL,
+        stdout=subprocess.PIPE,
+        stderr=subprocess.STDOUT,
+        universal_newlines=True,
+        encoding='utf-8',
+        errors='replace',
+        env=env,
+    )
+
+    # Write stdin data if provided, then close stdin
+    if stdin_data and process.stdin:
+        try:
+            process.stdin.write(stdin_data)
+            process.stdin.close()
+        except (BrokenPipeError, OSError):
+            pass
+
+    output_queue: queue.Queue[Optional[str]] = queue.Queue()
+    GRACEFUL_SHUTDOWN_DELAY = 0.3
+
+    def is_turn_completed(line: str) -> bool:
+        try:
+            data = json.loads(line)
+            return data.get("type") == "turn.completed"
+        except (json.JSONDecodeError, AttributeError, TypeError):
+            return False
+
+    def read_output() -> None:
+        if process.stdout:
+            for line in iter(process.stdout.readline, ""):
+                stripped = line.strip()
+                output_queue.put(stripped)
+                if is_turn_completed(stripped):
+                    time.sleep(GRACEFUL_SHUTDOWN_DELAY)
+                    process.terminate()
+                    break
+            process.stdout.close()
+        output_queue.put(None)
+
+    thread = threading.Thread(target=read_output)
+    thread.start()
+
+    while True:
+        try:
+            line = output_queue.get(timeout=0.5)
+            if line is None:
+                break
+            yield line
+        except queue.Empty:
+            if process.poll() is not None and not thread.is_alive():
+                break
+
+    try:
+        process.wait(timeout=5)
+    except subprocess.TimeoutExpired:
+        process.kill()
+        process.wait()
+    thread.join(timeout=5)
+
+    while not output_queue.empty():
+        try:
+            line = output_queue.get_nowait()
+            if line is not None:
+                yield line
+        except queue.Empty:
+            break
+
+def windows_escape(prompt):
+    """Windows style string escaping for newlines and special chars in prompt text."""
+    result = prompt.replace('\n', '\\n')
+    result = result.replace('\r', '\\r')
+    result = result.replace('\t', '\\t')
+    return result
+
+
+def configure_windows_stdio() -> None:
+    """Configure stdout/stderr to use UTF-8 encoding on Windows."""
+    if os.name != "nt":
+        return
+    for stream in (sys.stdout, sys.stderr):
+        reconfigure = getattr(stream, "reconfigure", None)
+        if callable(reconfigure):
+            try:
+                reconfigure(encoding="utf-8")
+            except (ValueError, OSError):
+                pass
+
+
+def main():
+    configure_windows_stdio()
+    parser = argparse.ArgumentParser(description="Codex Bridge")
+    parser.add_argument("--PROMPT", required=True, help="Instruction for the task to send to codex.")
+    parser.add_argument("--cd", required=True, help="Set the workspace root for codex before executing the task.")
+    parser.add_argument("--sandbox", default="read-only", choices=["read-only", "workspace-write", "danger-full-access"], help="Sandbox policy for model-generated commands. Defaults to `read-only`.")
+    parser.add_argument("--SESSION_ID", default="", help="Resume the specified session of the codex. Defaults to `None`, start a new session.")
+    parser.add_argument("--skip-git-repo-check", action="store_true", default=True, help="Allow codex running outside a Git repository (useful for one-off directories).")
+    parser.add_argument("--return-all-messages", action="store_true", help="Return all messages (e.g. reasoning, tool calls, etc.) from the codex session. Set to `False` by default, only the agent's final reply message is returned.")
+    parser.add_argument("--image", action="append", default=[], help="Attach one or more image files to the initial prompt. Separate multiple paths with commas or repeat the flag.")
+    parser.add_argument("--model", default="", help="The model to use for the codex session. This parameter is strictly prohibited unless explicitly specified by the user.")
+    parser.add_argument("--yolo", action="store_true", help="Run every command without approvals or sandboxing. Only use when `sandbox` couldn't be applied.")
+    parser.add_argument("--profile", default="", help="Configuration profile name to load from `~/.codex/config.toml`. This parameter is strictly prohibited unless explicitly specified by the user.")
+
+    args = parser.parse_args()
+
+    cmd = ["codex", "exec", "--sandbox", args.sandbox, "--cd", args.cd, "--json"]
+
+    if args.image:
+        cmd.extend(["--image", ",".join(args.image)])
+
+    if args.model:
+        cmd.extend(["--model", args.model])
+
+    if args.profile:
+        cmd.extend(["--profile", args.profile])
+
+    if args.yolo:
+        cmd.append("--yolo")
+
+    if args.skip_git_repo_check:
+        cmd.append("--skip-git-repo-check")
+
+    if args.SESSION_ID:
+        cmd.extend(["resume", args.SESSION_ID])
+
+    PROMPT = args.PROMPT
+
+    # Use stdin to pass PROMPT (use '-' as placeholder) to avoid shell argument parsing issues
+    cmd.append('-')
+
+    # Execution Logic
+    all_messages = []
+    agent_messages = ""
+    success = True
+    err_message = ""
+    thread_id = None
+
+    for line in run_shell_command(cmd, stdin_data=PROMPT):
+        try:
+            line_dict = json.loads(line.strip())
+            all_messages.append(line_dict)
+            item = line_dict.get("item", {})
+            item_type = item.get("type", "")
+            if item_type == "agent_message":
+                agent_messages = agent_messages + item.get("text", "")
+            if line_dict.get("thread_id") is not None:
+                thread_id = line_dict.get("thread_id")
+            if "fail" in line_dict.get("type", ""):
+                success = False if len(agent_messages) == 0 else success
+                err_message += "\n\n[codex error] " + line_dict.get("error", {}).get("message", "")
+            if "error" in line_dict.get("type", ""):
+                error_msg = line_dict.get("message", "")
+                is_reconnecting = bool(re.match(r'^Reconnecting\.\.\.\s+\d+/\d+$', error_msg))
+
+                if not is_reconnecting:
+                    success = False if len(agent_messages) == 0 else success
+                    err_message += "\n\n[codex error] " + error_msg
+
+        except json.JSONDecodeError:
+            err_message += "\n\n[json decode error] " + line
+            continue
+
+        except Exception as error:
+            err_message += "\n\n[unexpected error] " + f"Unexpected error: {error}. Line: {line!r}"
+            success = False
+            break
+
+    if thread_id is None:
+        success = False
+        err_message = "Failed to get `SESSION_ID` from the codex session. \n\n" + err_message
+
+    if len(agent_messages) == 0:
+        success = False
+        err_message = "Failed to get `agent_messages` from the codex session. \n\n You can try to set `return_all_messages` to `True` to get the full reasoning information. " + err_message
+
+    if success:
+        result = {
+            "success": True,
+            "SESSION_ID": thread_id,
+            "agent_messages": agent_messages,
+        }
+
+    else:
+        result = {"success": False, "error": err_message}
+
+    if args.return_all_messages:
+        result["all_messages"] = all_messages
+
+    print(json.dumps(result, indent=2, ensure_ascii=False))
+
+if __name__ == "__main__":
+    main()

package/.claude/skills/collaborating-with-gemini/SKILL.md

@@ -0,0 +1,194 @@
+---
+name: collaborating-with-gemini
+description: |
+  与 Google Gemini CLI 协同开发。将编码任务委托给 Gemini 进行前端原型、UI设计和代码审查。
+
+  触发场景：
+  - 需要前端/UI/样式原型设计
+  - 需要 CSS/React/Vue 组件设计
+  - 需要代码审查和 Bug 分析
+  - 用户明确要求使用 Gemini 协作
+  - 复杂前端逻辑的原型设计
+
+  触发词：Gemini、协作、多模型、前端原型、UI设计、CSS、样式、gemini协同
+
+  前置要求：
+  - 已安装 Gemini CLI (npm install -g @google/gemini-cli)
+  - 已配置 Google API Key
+
+  注意：Gemini 对后端逻辑理解有缺陷，后端任务优先使用 Codex。
+---
+
+# 与 Gemini CLI 协同开发
+
+> 通过 Python 桥接脚本调用 Gemini CLI，获取前端原型和 UI 设计建议。
+
+## 快速开始
+
+```bash
+# 相对路径（推荐，在项目根目录执行）
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py --cd . --PROMPT "Your task"
+```
+
+**输出**: JSON 格式，包含 `success`、`SESSION_ID`、`agent_messages` 和可选的 `error`。
+
+## 参数说明
+
+| 参数 | 类型 | 必填 | 默认值 | 说明 |
+|------|------|------|--------|------|
+| `--PROMPT` | str | ✅ | - | 发送给 Gemini 的任务指令（使用英语） |
+| `--cd` | Path | ✅ | - | 工作目录根路径 |
+| `--sandbox` | bool | ❌ | `False` | 是否启用沙箱模式 |
+| `--SESSION_ID` | str | ❌ | `""` | 会话 ID（继续之前的对话） |
+| `--return-all-messages` | bool | ❌ | `False` | 返回完整推理信息 |
+| `--model` | str | ❌ | `None` | 指定模型（仅用户明确要求时使用） |
+
+## 使用模式
+
+### 1. 基础调用
+
+```bash
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py \
+  --cd . \
+  --PROMPT "Design a responsive card component for product display"
+```
+
+### 2. 多轮会话
+
+**始终保存 SESSION_ID** 用于后续对话：
+
+```bash
+# 第一轮：设计 UI
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py \
+  --cd "/project" \
+  --PROMPT "Design a mobile-first login page with form validation"
+
+# 后续轮次：使用 SESSION_ID 继续
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py \
+  --cd "/project" \
+  --SESSION_ID "uuid-from-previous-response" \
+  --PROMPT "Add dark mode support to the login page design"
+```
+
+### 3. 获取 Unified Diff 补丁
+
+```bash
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py \
+  --cd "/project" \
+  --PROMPT "Generate a unified diff to improve the CSS layout of Home.vue. OUTPUT: Unified Diff Patch ONLY."
+```
+
+### 4. 调试模式（返回完整信息）
+
+```bash
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py \
+  --cd "/project" \
+  --PROMPT "Debug this styling issue: elements overflow on mobile" \
+  --return-all-messages
+```
+
+## 返回值结构
+
+**成功时：**
+```json
+{
+  "success": true,
+  "SESSION_ID": "550e8400-e29b-41d4-a716-446655440000",
+  "agent_messages": "模型回复内容..."
+}
+```
+
+**失败时：**
+```json
+{
+  "success": false,
+  "error": "错误信息描述"
+}
+```
+
+## 协作工作流
+
+### 推荐场景
+
+| 场景 | 说明 |
+|------|------|
+| **前端/UI/UX** | Gemini 擅长 CSS、样式和视觉设计 |
+| **组件设计** | React/Vue 组件的原型设计 |
+| **响应式布局** | 移动端适配和布局优化 |
+| **样式审查** | CSS 代码质量和最佳实践 |
+
+### 重要约束
+
+1. **上下文限制**: Gemini 上下文 < 32k tokens，避免过长输入
+2. **英语交互**: 与 Gemini 交互时使用英语，获得更好效果
+3. **中文输出强制**: Gemini 倾向于输出英文，必须在每次 PROMPT 末尾追加以下约束：
+   ```
+   IMPORTANT LANGUAGE RULES:
+   - All SQL COMMENT values MUST be in Chinese (e.g., COMMENT '用户名' NOT COMMENT 'username')
+   - All Java/code comments (Javadoc, inline //, block /* */) MUST be in Chinese
+   - All field descriptions, table descriptions MUST be in Chinese
+   - Variable names and class names remain in English (camelCase/PascalCase)
+   ```
+4. **Diff 输出**: 在 PROMPT 中明确要求 `OUTPUT: Unified Diff Patch ONLY`
+5. **后端逻辑**: Gemini 对后端逻辑理解有缺陷，后端任务优先使用 Codex
+6. **重构代码**: 将 Gemini 的输出视为"脏原型"，由 Claude 重构为生产代码（**特别检查注释和 COMMENT 是否为中文，不是则修正**）
+7. **后台运行**: 对于长时间任务，使用 `Run in the background`
+
+## 与本项目的集成
+
+### 典型用例：后端代码审查
+
+```bash
+# 审查后端 Controller 代码
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py \
+  --cd . \
+  --PROMPT "Review the RESTful API design in ruoyi-modules/ruoyi-system/src/main/java/org/dromara/system/controller/. Focus on path conventions (/list, /{id}) and proper use of @SaCheckPermission."
+```
+
+### 典型用例：Service 层代码审查
+
+```bash
+# 审查 Service 实现
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py \
+  --cd "/project" \
+  --PROMPT "Review the XxxServiceImpl.java for buildQueryWrapper implementation, MapstructUtils usage, and adherence to three-layer architecture (no DAO layer). OUTPUT: Review comments."
+```
+
+### 典型用例：Entity 设计审查
+
+```bash
+# 审查 Entity 设计
+python .claude/skills/collaborating-with-gemini/scripts/gemini_bridge.py \
+  --cd "/project" \
+  --PROMPT "Review the Entity classes in ruoyi-modules/ruoyi-demo/src/main/java/org/dromara/demo/domain/ for proper TenantEntity inheritance, @TableLogic usage, and field naming conventions."
+```
+
+## 安装前置
+
+```bash
+# 安装 Gemini CLI
+npm install -g @google/gemini-cli
+
+# 配置 API Key
+gemini auth login
+```
+
+## 故障排除
+
+| 问题 | 解决方案 |
+|------|---------|
+| `gemini: command not found` | 确保已安装并添加到 PATH |
+| `SESSION_ID` 获取失败 | 检查网络连接和 API Key |
+| 输出被截断 | 使用 `--return-all-messages` 获取完整信息 |
+| 上下文过长 | 减少输入内容，分批处理 |
+| Windows 路径问题 | 使用正斜杠 `/` 或双反斜杠 `\\` |
+
+## Gemini vs Codex 选择指南
+
+| 任务类型 | 推荐模型 | 原因 |
+|---------|---------|------|
+| 前端 UI/CSS | Gemini | 视觉设计能力强 |
+| 后端逻辑 | Codex | 算法和逻辑分析强 |
+| 组件样式 | Gemini | CSS 和布局专长 |
+| API 设计 | Codex | 接口设计和架构 |
+| 代码审查 | 两者皆可 | 双模型交叉验证更好 |