sumulige-claude 1.0.0

package/template/.claude/boris-optimizations.md

@@ -0,0 +1,167 @@
+# Boris 优化指南
+
+本指南基于 Claude Code 创造者 Boris Cherny 的最佳实践，为 oh-my-claude 项目提供优化建议。
+
+## 已实现的优化
+
+### 1. 验证反馈循环 🔍
+
+**文件**: `.claude/hooks/verify-work.cjs`
+
+**重要性**: ⭐⭐⭐⭐⭐ (最重要！质量提升 2-3 倍)
+
+**功能**:
+- 在关键操作后自动提醒验证
+- 记录待验证任务列表
+- 生成验证日志
+
+**使用方式**:
+```bash
+# 手动触发验证流程
+/verify-work
+```
+
+**验证时机**:
+- git commit 后 - 运行测试确保功能正常
+- git push 后 - 检查 CI/CD 状态
+- deploy 后 - 在预发布环境验证
+- build 后 - 检查构建产物
+
+---
+
+### 2. Slash Commands ⚡
+
+**目录**: `.claude/commands/`
+
+**可用命令**:
+
+| 命令 | 功能 |
+|------|------|
+| `/commit-push-pr` | 提交代码、推送、创建 PR |
+| `/commit` | 创建 git commit |
+| `/pr` | 创建或更新 PR |
+| `/test` | 运行测试套件 |
+| `/review` | 审查当前更改 |
+| `/verify-work` | 查看待验证任务 |
+| `/sessions` | 管理并行会话 |
+
+**使用方式**:
+```bash
+# 快速提交并创建 PR
+/commit-push-pr
+
+# 审查更改后再提交
+/review
+/commit
+
+# 运行测试
+/test
+```
+
+---
+
+### 3. 代码格式化 Hook 🎨
+
+**文件**: `.claude/hooks/code-formatter.cjs`
+
+**功能**:
+- 在代码编辑后自动格式化
+- 支持多种语言和格式化工具
+- 静默运行，不打扰工作流
+
+**支持的语言**:
+- JavaScript/TypeScript (Prettier)
+- Python (Black)
+- Rust (rustfmt)
+- Go (gofmt)
+- JSON/YAML/Markdown (Prettier)
+
+**调试模式**:
+```bash
+export DEBUG_FORMATTER=1
+```
+
+---
+
+### 4. 并行多会话支持 🔄
+
+**文件**: `.claude/hooks/multi-session.cjs`
+
+**功能**:
+- 跟踪活跃的并行会话
+- 为每个会话分配编号 (1-10)
+- 自动清理过期会话
+
+**Boris 的工作流**:
+```
+终端标签 1: Conductor (协调/规划)
+终端标签 2: Architect (设计/架构)
+终端标签 3: Builder (实现)
+终端标签 4: Reviewer (审查)
+终端标签 5: Explorer (研究)
+
++ 5-10 个 claude.ai/code Web 会话
+```
+
+**使用方式**:
+```bash
+# 查看所有活跃会话
+/sessions
+
+# 或直接运行
+node .claude/hooks/multi-session.cjs --status
+```
+
+---
+
+## 配置更新
+
+所有 hooks 已添加到 `.claude/settings.json`:
+
+```json
+{
+  "matcher": "UserPromptSubmit|PreToolUse|PostToolUse|AgentStop",
+  "hooks": [
+    { "command": "...project-kickoff.cjs", "timeout": 1 },
+    { "command": "...thinking-silent.cjs", "timeout": 1 },
+    { "command": "...rag-skill-loader.cjs", "timeout": 1 },
+    { "command": "...code-formatter.cjs", "timeout": 5 },
+    { "command": "...verify-work.cjs", "timeout": 1 },
+    { "command": "...multi-session.cjs", "timeout": 1 }
+  ]
+}
+```
+
+---
+
+## 最佳实践总结
+
+### 启动会话
+1. **Plan 模式开始** (Shift+Tab 两次)
+2. 批准计划后切换到 auto-accept 模式
+3. 使用 `/commit-push-pr` 完成工作流
+
+### 并行工作
+1. 开启 5 个终端标签页
+2. 每个标签运行不同的 agent/任务
+3. 使用系统通知跟踪进度
+4. 同时使用 claude.ai/code Web 会话
+
+### 验证工作
+1. 每次提交前 `/review`
+2. 提交后 `/test` 验证
+3. 定期 `/verify-work` 查看待验证项
+4. 使用 background agent 验证长时间任务
+
+### 团队协作
+1. 共享 CLAUDE.md 到 git
+2. PR 审查时使用 @.claude 添加规范
+3. 定期更新团队知识库
+
+---
+
+## 额外资源
+
+- [Claude Code 文档](https://code.claude.com)
+- [Boris 的 Twitter Thread](https://twitter-thread.com/t/2007179832300581177)
+- [Claude Chrome Extension](https://code.claude.com/docs/en/chrome)

package/template/.claude/commands/commit-push-pr.md

@@ -0,0 +1,59 @@
+---
+description: Create git commit, push to remote, and create/merge pull request
+---
+
+You are tasked with creating a git commit, pushing to remote, and handling the pull request workflow.
+
+## Step 1: Gather Information
+
+First, run these commands in parallel to understand the current state:
+
+```bash
+git status
+git diff --staged
+git log -3 --oneline
+git diff HEAD~3..HEAD --stat
+```
+
+## Step 2: Analyze and Plan
+
+Analyze the changes and:
+1. Determine what type of change this is (feat, fix, docs, refactor, test, chore)
+2. Identify the main purpose of the changes
+3. Check if there are any sensitive files that shouldn't be committed
+
+## Step 3: Stage and Commit
+
+If there are unstaged changes that should be included, ask the user first.
+
+Then create a commit with a clear message following this format:
+
+```
+<type>: <brief description>
+
+<optional detailed explanation>
+
+Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
+```
+
+## Step 4: Push and Create PR
+
+1. Push the changes to remote
+2. Check if there's an open PR for this branch using:
+   ```bash
+   gh pr list --head $(git branch --show-current)
+   ```
+3. If PR exists, ask if user wants to update it
+4. If no PR, create one using `gh pr create` with:
+   - A clear title based on the commit message
+   - A body with:
+     - Summary of changes (2-3 bullet points)
+     - Test plan (checklist)
+     - Link to any related issues
+
+## Step 5: Verify
+
+After creating the PR, show the PR URL and check CI status using:
+```bash
+gh pr checks
+```

package/template/.claude/commands/commit.md

@@ -0,0 +1,53 @@
+---
+description: Create a git commit with proper message format
+---
+
+Create a git commit following best practices.
+
+## Step 1: Assess Current State
+
+Run these commands in parallel:
+
+```bash
+git status
+git diff
+git log -5 --oneline
+```
+
+## Step 2: Stage Changes
+
+1. Review what files are changed
+2. Ask user if all changes should be committed, or if some should be staged separately
+3. Never commit these patterns:
+   - `.env`, `.env.local`, credentials
+   - `node_modules/`, `dist/`, `build/`
+   - `.DS_Store`, `*.log`
+   - Any API keys, tokens, or secrets
+
+## Step 3: Write Commit Message
+
+Follow this format based on change type:
+
+| Type | Usage |
+|------|-------|
+| `feat` | New feature |
+| `fix` | Bug fix |
+| `docs` | Documentation only |
+| `style` | Code style changes (formatting) |
+| `refactor` | Code refactoring |
+| `test` | Adding or updating tests |
+| `chore` | Build process, tooling |
+
+Commit message template:
+
+```
+<type>: <brief description (50 chars max)>
+
+<detailed explanation if needed>
+
+Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
+```
+
+## Step 4: Create Commit
+
+Use git commit with the message, then run `git status` to confirm.

package/template/.claude/commands/pr.md

@@ -0,0 +1,76 @@
+---
+description: Create or update a GitHub pull request
+---
+
+Create or update a pull request with proper documentation.
+
+## Step 1: Check Current State
+
+```bash
+git status
+git log -5 --oneline
+git branch --show-current
+```
+
+## Step 2: Check for Existing PR
+
+```bash
+gh pr list --head $(git branch --show-current)
+```
+
+## Step 3: Create PR (if none exists)
+
+### Analyze Changes
+
+```bash
+# Get commit history since branch from main/master
+git diff $(git merge-base main HEAD)..HEAD --stat
+
+# Get detailed diff
+git diff $(git merge-base main HEAD)..HEAD
+```
+
+### Create PR with Template
+
+Use `gh pr create` with:
+
+```markdown
+## Summary
+
+<!-- 2-3 bullet points describing the main changes -->
+
+## Changes
+
+<!-- Detailed list of changes -->
+
+## Test Plan
+
+- [ ] Unit tests pass
+- [ ] Integration tests pass
+- [ ] Manual testing completed
+- [ ] Documentation updated
+
+## Checklist
+
+- [ ] Code follows project style guidelines
+- [ ] Self-review completed
+- [ ] Comments added to complex code
+- [ ] Documentation updated
+- [ ] No new warnings generated
+```
+
+## Step 4: Update PR (if exists)
+
+If PR already exists, ask user if they want to:
+- Update description
+- Add comments
+- Request review
+- Merge if ready
+
+## Step 5: Show Result
+
+Display PR URL and check CI status:
+
+```bash
+gh pr checks
+```

package/template/.claude/commands/review.md

@@ -0,0 +1,61 @@
+---
+description: Review current changes before committing
+---
+
+Review the current uncommitted or staged changes.
+
+## Step 1: Show Changes
+
+```bash
+git status
+
+# Show unstaged changes
+git diff
+
+# Show staged changes
+git diff --staged
+```
+
+## Step 2: Code Review Checklist
+
+Review the changes for:
+
+### Functionality
+- [ ] Does the change accomplish the intended goal?
+- [ ] Are there any obvious bugs or logic errors?
+- [ ] Are edge cases handled?
+
+### Security
+- [ ] No hardcoded credentials or API keys
+- [ ] No SQL injection, XSS, or other vulnerabilities
+- [ ] Proper input validation
+
+### Code Quality
+- [ ] Code is readable and self-documenting
+- [ ] No unnecessary complexity
+- [ ] Follows DRY (Don't Repeat Yourself)
+- [ ] Proper error handling
+
+### Testing
+- [ ] Tests cover new functionality
+- [ ] No tests are broken
+- [ ] Edge cases are tested
+
+### Performance
+- [ ] No obvious performance issues
+- [ ] No unnecessary database queries or API calls
+- [ ] Proper use of caching (if applicable)
+
+## Step 3: Provide Feedback
+
+For each issue found:
+1. Point out the specific file and line
+2. Explain the issue
+3. Suggest a fix
+
+## Step 4: Summary
+
+Provide an overall assessment:
+- ✅ Ready to commit
+- ⚠️ Minor issues to address
+- ❌ Major issues that need fixing

package/template/.claude/commands/sessions.md

@@ -0,0 +1,62 @@
+---
+description: Manage and display parallel Claude sessions
+---
+
+Manage and display information about parallel Claude sessions.
+
+## Show Session Status
+
+Display all active parallel sessions:
+
+```bash
+node .claude/hooks/multi-session.cjs --status
+```
+
+## Session Management Best Practices
+
+Based on Boris Cherny's workflow:
+
+### Terminal Sessions (Local)
+- Number your terminal tabs 1-5
+- Use system notifications to know when Claude needs input
+- Run different agents in different tabs:
+  - Tab 1: Conductor (planning/coordination)
+  - Tab 2: Architect (design/architecture)
+  - Tab 3: Builder (implementation)
+  - Tab 4: Reviewer (code review)
+  - Tab 5: Explorer (research)
+
+### Web Sessions (claude.ai/code)
+- Run 5-10 web sessions in parallel with local sessions
+- Use `&` to hand off local sessions to web
+- Use `--teleport` to switch between local and web
+- Start sessions from mobile Claude app for async work
+
+### Session Handoff
+
+To hand off a session to web:
+```bash
+# End local session with summary
+# The session context will be available in claude.ai/code
+```
+
+## View Session History
+
+```bash
+cat .claude/sessions/active-sessions.json
+```
+
+## Clean Old Sessions
+
+Sessions older than 1 hour are automatically cleaned up. To manually clean:
+
+```bash
+rm .claude/sessions/active-sessions.json
+```
+
+## Tips
+
+1. **Use Plan mode** (Shift+Tab twice) for most sessions
+2. **Switch to auto-accept** after approving a plan
+3. **Let sessions run in parallel** - don't wait for one to finish
+4. **Check notifications** to see when sessions need input

package/template/.claude/commands/skill-create.md

@@ -0,0 +1,131 @@
+---
+description: Create a new skill with template
+---
+
+Create a new skill using the standard template.
+
+## Step 1: Get Skill Information
+
+Ask the user for:
+1. **Skill name** (kebab-case, e.g., `code-reviewer`)
+2. **Description** (one-line summary)
+3. **Category** (development, design, analysis, workflow, etc.)
+4. **Difficulty** (beginner, intermediate, advanced)
+
+## Step 2: Create Skill Directory
+
+Create the skill directory structure:
+
+```bash
+mkdir -p .claude/skills/{skill-name}/{templates,examples}
+```
+
+## Step 3: Generate SKILL.md
+
+Use this template:
+
+```markdown
+# {Skill Name}
+
+> {Description}
+
+**版本**: 1.0.0
+**作者**: AI
+**标签**: [{category}]
+**难度**: {difficulty}
+
+---
+
+## 概述
+
+{Detailed description of what this skill does}
+
+## 适用场景
+
+- {Scenario 1}
+- {Scenario 2}
+- {Scenario 3}
+
+## 触发关键词
+
+```
+{trigger keywords}
+```
+
+## 使用方法
+
+### 基础用法
+
+{Example usage}
+
+## 输出格式
+
+{Output format description}
+
+## 注意事项
+
+- {Note 1}
+- {Note 2}
+
+## 相关技能
+
+- [Related Skill](../related-skill/)
+
+## 更新日志
+
+### 1.0.0 ({current-date})
+- 初始版本
+```
+
+## Step 4: Generate metadata.yaml
+
+```yaml
+name: {skill-name}
+version: 1.0.0
+author: AI
+description: {description}
+
+tags:
+  - {category}
+
+triggers:
+  - {trigger1}
+  - {trigger2}
+
+dependencies: []
+difficulty: {difficulty}
+```
+
+## Step 5: Add to RAG Index
+
+Update `.claude/rag/skill-index.json`:
+
+```json
+{
+  "skills": [
+    {
+      "name": "{skill-name}",
+      "description": "{description}",
+      "keywords": ["{trigger1}", "{trigger2}"],
+      "path": ".claude/skills/{skill-name}/SKILL.md"
+    }
+  ]
+}
+```
+
+## Step 6: Verify
+
+Confirm with user:
+- Skill created at `.claude/skills/{skill-name}/`
+- Added to RAG index
+- Ready to use
+
+## Example
+
+User: "Create a skill called api-tester for testing REST APIs"
+
+→ Create `.claude/skills/api-tester/` with:
+- SKILL.md (filled with API testing content)
+- metadata.yaml
+- templates/request.md
+- examples/basic-test.md

package/template/.claude/commands/test.md

@@ -0,0 +1,56 @@
+---
+description: Run tests and show coverage report
+---
+
+Run the test suite and provide comprehensive results.
+
+## Step 1: Detect Test Framework
+
+Check what test framework is being used:
+
+```bash
+cat package.json | grep -E '"test"|"jest"|"vitest"|"mocha"|"pytest"'
+```
+
+## Step 2: Run Tests
+
+Run the appropriate test command:
+
+```bash
+# Node.js / JavaScript
+npm test
+
+# Python
+pytest
+
+# Rust
+cargo test
+
+# Go
+go test ./...
+```
+
+## Step 3: Show Coverage
+
+If coverage is available, run:
+
+```bash
+npm run test:coverage
+# or
+pytest --cov
+```
+
+## Step 4: Analyze Results
+
+Provide a summary:
+- Total tests run
+- Pass/fail counts
+- Coverage percentage
+- Any failing tests with details
+
+## Step 5: Fix Failures (if any)
+
+If tests fail, help the user:
+1. Identify the root cause
+2. Fix the issue
+3. Re-run tests to verify