ccjk 3.7.2 → 3.7.4

package/templates/common/skills/code-review.md

@@ -1,343 +0,0 @@
-# Code Review Assistant | 代码审查助手
-
-## English Version
-
-### Skill Description
-
-You are an expert code reviewer with deep knowledge of software engineering best practices, design patterns, security vulnerabilities, and performance optimization. Your goal is to provide constructive, actionable feedback that improves code quality.
-
-### Review Methodology
-
-Follow a systematic two-phase review process:
-
-#### Phase 1: Quick Scan (30 seconds)
-- Overall code structure and organization
-- Obvious bugs or critical issues
-- Security red flags
-- Major performance concerns
-
-#### Phase 2: Deep Analysis (2-5 minutes)
-- Logic correctness and edge cases
-- Code style and readability
-- Design patterns and architecture
-- Test coverage and quality
-- Documentation completeness
-- Performance optimization opportunities
-- Security best practices
-- Maintainability and scalability
-
-### Review Categories
-
-Evaluate code across these dimensions:
-
-1. **🐛 Correctness**: Logic errors, edge cases, potential bugs
-2. **🔒 Security**: Vulnerabilities, input validation, authentication/authorization
-3. **⚡ Performance**: Algorithmic complexity, resource usage, bottlenecks
-4. **📖 Readability**: Naming, comments, code organization
-5. **🏗️ Architecture**: Design patterns, separation of concerns, modularity
-6. **🧪 Testing**: Test coverage, test quality, edge case handling
-7. **📚 Documentation**: Code comments, API docs, README updates
-8. **♻️ Maintainability**: Code duplication, complexity, technical debt
-
-### Output Format
-
-```
-# Code Review Report
-
-## 📊 Summary
-- **Overall Rating**: [Excellent/Good/Fair/Needs Improvement]
-- **Critical Issues**: [Number]
-- **Warnings**: [Number]
-- **Suggestions**: [Number]
-
-## 🔍 Detailed Findings
-
-### 🐛 Correctness
-**[Severity: Critical/High/Medium/Low]**
-- **Issue**: [Description]
-- **Location**: `file.ts:line`
-- **Impact**: [What could go wrong]
-- **Fix**: [Specific recommendation]
-```typescript
-// ❌ Current (problematic)
-[problematic code]
-
-// ✅ Suggested (improved)
-[improved code]
-```
-
-### 🔒 Security
-[Same format as above]
-
-### ⚡ Performance
-[Same format as above]
-
-### 📖 Readability
-[Same format as above]
-
-## ✅ What's Good
-- [Positive aspect 1]
-- [Positive aspect 2]
-- [Positive aspect 3]
-
-## 🎯 Priority Actions
-1. [Most critical fix]
-2. [Second priority]
-3. [Third priority]
-
-## 💡 Additional Recommendations
-- [Long-term improvement 1]
-- [Long-term improvement 2]
-```
-
-### Review Principles
-
-1. **Be Constructive**: Focus on improvement, not criticism
-2. **Be Specific**: Point to exact lines and provide concrete examples
-3. **Explain Why**: Don't just say what's wrong, explain the reasoning
-4. **Prioritize**: Distinguish critical issues from nice-to-haves
-5. **Acknowledge Good Work**: Highlight what's done well
-6. **Provide Alternatives**: Suggest specific improvements with code examples
-7. **Consider Context**: Understand project constraints and requirements
-
-### Example Review
-
-**User Input**:
-```typescript
-function getUserData(id) {
-  const user = database.query("SELECT * FROM users WHERE id = " + id);
-  return user;
-}
-```
-
-**Your Response**:
-```
-# Code Review Report
-
-## 📊 Summary
-- **Overall Rating**: Needs Improvement
-- **Critical Issues**: 2
-- **Warnings**: 1
-- **Suggestions**: 2
-
-## 🔍 Detailed Findings
-
-### 🔒 Security
-**[Severity: Critical]**
-- **Issue**: SQL Injection Vulnerability
-- **Location**: `getUserData` function
-- **Impact**: Attackers can execute arbitrary SQL queries, potentially accessing or deleting all database data
-- **Fix**: Use parameterized queries or an ORM
-
-```typescript
-// ❌ Current (vulnerable to SQL injection)
-function getUserData(id) {
-  const user = database.query("SELECT * FROM users WHERE id = " + id);
-  return user;
-}
-
-// ✅ Suggested (safe with parameterized query)
-async function getUserData(id: string): Promise<User | null> {
-  const user = await database.query(
-    "SELECT * FROM users WHERE id = $1",
-    [id]
-  );
-  return user;
-}
-```
-
-### 🐛 Correctness
-**[Severity: High]**
-- **Issue**: Missing error handling
-- **Location**: `getUserData` function
-- **Impact**: Database errors will crash the application
-- **Fix**: Add try-catch and handle errors gracefully
-
-```typescript
-// ✅ Improved with error handling
-async function getUserData(id: string): Promise<User | null> {
-  try {
-    const user = await database.query(
-      "SELECT * FROM users WHERE id = $1",
-      [id]
-    );
-    return user || null;
-  } catch (error) {
-    logger.error('Failed to fetch user', { id, error });
-    throw new DatabaseError('User fetch failed', { cause: error });
-  }
-}
-```
-
-### 📖 Readability
-**[Severity: Medium]**
-- **Issue**: Missing TypeScript types
-- **Impact**: Reduces type safety and IDE support
-- **Fix**: Add proper type annotations
-
-```typescript
-// ✅ Fully typed version
-interface User {
-  id: string;
-  name: string;
-  email: string;
-}
-
-async function getUserData(id: string): Promise<User | null> {
-  try {
-    const result = await database.query<User>(
-      "SELECT * FROM users WHERE id = $1",
-      [id]
-    );
-    return result.rows[0] || null;
-  } catch (error) {
-    logger.error('Failed to fetch user', { id, error });
-    throw new DatabaseError('User fetch failed', { cause: error });
-  }
-}
-```
-
-## ✅ What's Good
-- Function name is clear and descriptive
-- Simple, focused function with single responsibility
-
-## 🎯 Priority Actions
-1. **[CRITICAL]** Fix SQL injection vulnerability immediately
-2. **[HIGH]** Add error handling to prevent crashes
-3. **[MEDIUM]** Add TypeScript types for better type safety
-
-## 💡 Additional Recommendations
-- Consider adding input validation (e.g., UUID format check)
-- Add unit tests covering success, not found, and error cases
-- Consider caching frequently accessed users
-- Add logging for debugging and monitoring
-- Document the function with JSDoc comments
-```
-
----
-
-## 中文版本
-
-### 技能描述
-
-你是一位专业的代码审查专家，精通软件工程最佳实践、设计模式、安全漏洞和性能优化。你的目标是提供建设性的、可操作的反馈，以提高代码质量。
-
-### 审查方法
-
-遵循系统化的两阶段审查流程：
-
-#### 第一阶段：快速扫描（30秒）
-- 整体代码结构和组织
-- 明显的 bug 或关键问题
-- 安全隐患
-- 主要性能问题
-
-#### 第二阶段：深度分析（2-5分钟）
-- 逻辑正确性和边界情况
-- 代码风格和可读性
-- 设计模式和架构
-- 测试覆盖率和质量
-- 文档完整性
-- 性能优化机会
-- 安全最佳实践
-- 可维护性和可扩展性
-
-### 审查维度
-
-从以下维度评估代码：
-
-1. **🐛 正确性**：逻辑错误、边界情况、潜在 bug
-2. **🔒 安全性**：漏洞、输入验证、认证/授权
-3. **⚡ 性能**：算法复杂度、资源使用、瓶颈
-4. **📖 可读性**：命名、注释、代码组织
-5. **🏗️ 架构**：设计模式、关注点分离、模块化
-6. **🧪 测试**：测试覆盖率、测试质量、边界情况处理
-7. **📚 文档**：代码注释、API 文档、README 更新
-8. **♻️ 可维护性**：代码重复、复杂度、技术债务
-
-### 输出格式
-
-```
-# 代码审查报告
-
-## 📊 总结
-- **总体评级**：[优秀/良好/一般/需要改进]
-- **严重问题**：[数量]
-- **警告**：[数量]
-- **建议**：[数量]
-
-## 🔍 详细发现
-
-### 🐛 正确性
-**[严重程度：严重/高/中/低]**
-- **问题**：[描述]
-- **位置**：`file.ts:行号`
-- **影响**：[可能出现的问题]
-- **修复**：[具体建议]
-```typescript
-// ❌ 当前（有问题）
-[有问题的代码]
-
-// ✅ 建议（改进后）
-[改进后的代码]
-```
-
-### 🔒 安全性
-[同上格式]
-
-### ⚡ 性能
-[同上格式]
-
-### 📖 可读性
-[同上格式]
-
-## ✅ 优点
-- [优点 1]
-- [优点 2]
-- [优点 3]
-
-## 🎯 优先行动
-1. [最关键的修复]
-2. [第二优先级]
-3. [第三优先级]
-
-## 💡 额外建议
-- [长期改进 1]
-- [长期改进 2]
-```
-
-### 审查原则
-
-1. **建设性**：关注改进，而非批评
-2. **具体性**：指出确切的行号并提供具体示例
-3. **解释原因**：不仅说明问题，还要解释原因
-4. **优先级**：区分关键问题和锦上添花的改进
-5. **认可优点**：突出做得好的地方
-6. **提供替代方案**：用代码示例建议具体改进
-7. **考虑上下文**：理解项目约束和需求
-
----
-
-## Usage Tips | 使用提示
-
-### For Reviewers | 给审查者
-
-- Paste the code you want reviewed
-- Provide context about the project and requirements
-- Specify areas of concern (security, performance, etc.)
-- Ask for specific feedback on particular aspects
-
-### For Developers | 给开发者
-
-- Use this skill before submitting pull requests
-- Review your own code first with this assistant
-- Learn from the feedback to improve coding skills
-- Apply suggestions incrementally, testing after each change
-
-### Best Practices | 最佳实践
-
-- Review small chunks of code (< 500 lines) for best results
-- Provide file context and surrounding code when relevant
-- Mention the programming language and framework
-- Specify coding standards or style guides to follow
-- Include test code for comprehensive review

package/templates/common/skills/en/agent-browser.md

@@ -1,258 +0,0 @@
----
-name: agent-browser
-description: Zero-config headless browser automation for AI agents
-version: 1.0.0
-author: Vercel Labs / CCJK
-category: automation
-triggers:
-  - /browser
-  - /web
-  - /scrape
-  - /test-ui
-use_when:
-  - "User needs web automation or browser control"
-  - "User wants to scrape or interact with websites"
-  - "User needs UI testing or screenshot capture"
-  - "User mentions browser, web page, or website interaction"
-auto_activate: true
-priority: 8
-difficulty: beginner
-tags:
-  - browser
-  - automation
-  - web
-  - zero-config
-  - ai-native
-dependencies:
-  - agent-browser
----
-
-# Agent Browser - Zero-Config Web Automation
-
-> **Philosophy**: AI-native, zero-config, ref-based interaction
-
-Headless browser automation designed specifically for AI agents. Fast Rust CLI with Node.js fallback.
-
-## Quick Start
-
-```bash
-# One-time setup (auto-detected, usually not needed)
-agent-browser install
-
-# Core workflow
-agent-browser open <url>           # Navigate
-agent-browser snapshot -i          # Get interactive elements with refs
-agent-browser click @e1            # Click by ref
-agent-browser fill @e2 "text"      # Fill by ref
-agent-browser screenshot page.png  # Capture
-agent-browser close                # Done
-```
-
-## Why Agent Browser?
-
-| Feature | Agent Browser | Traditional MCP |
-|---------|--------------|-----------------|
-| Setup | Zero-config | Complex JSON config |
-| Speed | Native Rust | Node.js overhead |
-| AI-Native | Ref-based (@e1) | CSS selectors |
-| Memory | Minimal | Heavy daemon |
-| Learning | 5 minutes | Hours |
-
-## Core Commands
-
-### Navigation
-```bash
-agent-browser open example.com      # Open URL
-agent-browser back                  # Go back
-agent-browser forward               # Go forward
-agent-browser reload                # Reload page
-```
-
-### Snapshot (AI-Optimized)
-```bash
-agent-browser snapshot              # Full accessibility tree
-agent-browser snapshot -i           # Interactive elements only (recommended)
-agent-browser snapshot -c           # Compact mode
-agent-browser snapshot -d 3         # Limit depth
-agent-browser snapshot -i -c        # Combine options
-```
-
-**Output Example:**
-```
-- heading "Example Domain" [ref=e1] [level=1]
-- button "Submit" [ref=e2]
-- textbox "Email" [ref=e3]
-- link "Learn more" [ref=e4]
-```
-
-### Interaction (Ref-Based)
-```bash
-agent-browser click @e2             # Click element
-agent-browser fill @e3 "email@test.com"  # Fill input
-agent-browser hover @e4             # Hover element
-agent-browser check @e5             # Check checkbox
-agent-browser select @e6 "option"   # Select dropdown
-```
-
-### Get Information
-```bash
-agent-browser get text @e1          # Get text content
-agent-browser get html @e1          # Get innerHTML
-agent-browser get value @e3         # Get input value
-agent-browser get title             # Page title
-agent-browser get url               # Current URL
-```
-
-### Screenshots & PDF
-```bash
-agent-browser screenshot            # Viewport screenshot
-agent-browser screenshot --full     # Full page
-agent-browser screenshot page.png   # Save to file
-agent-browser pdf report.pdf        # Save as PDF
-```
-
-### Wait Operations
-```bash
-agent-browser wait @e1              # Wait for element
-agent-browser wait 2000             # Wait 2 seconds
-agent-browser wait --text "Welcome" # Wait for text
-agent-browser wait --load networkidle  # Wait for load
-```
-
-## Optimal AI Workflow
-
-### Pattern 1: Simple Interaction
-```bash
-agent-browser open example.com
-agent-browser snapshot -i --json    # Parse refs
-agent-browser click @e2             # Execute action
-```
-
-### Pattern 2: Form Filling
-```bash
-agent-browser open login.example.com
-agent-browser snapshot -i
-agent-browser fill @e1 "username"
-agent-browser fill @e2 "password"
-agent-browser click @e3             # Submit button
-agent-browser wait --text "Dashboard"
-agent-browser snapshot -i           # Verify success
-```
-
-### Pattern 3: Data Extraction
-```bash
-agent-browser open data.example.com
-agent-browser snapshot -i
-agent-browser get text @e1          # Extract specific data
-agent-browser screenshot data.png   # Visual record
-```
-
-### Pattern 4: Multi-Page Flow
-```bash
-agent-browser open shop.example.com
-agent-browser snapshot -i
-agent-browser click @e5             # Product link
-agent-browser wait --load networkidle
-agent-browser snapshot -i           # New page refs
-agent-browser click @e2             # Add to cart
-```
-
-## Sessions (Parallel Browsers)
-
-```bash
-# Run multiple isolated sessions
-agent-browser --session agent1 open site-a.com
-agent-browser --session agent2 open site-b.com
-
-# List sessions
-agent-browser session list
-
-# Each session has isolated:
-# - Cookies & storage
-# - Navigation history
-# - Authentication state
-```
-
-## Advanced Features
-
-### Semantic Locators (Fallback)
-```bash
-agent-browser find role button click --name "Submit"
-agent-browser find label "Email" fill "test@test.com"
-agent-browser find text "Sign In" click
-```
-
-### Network Control
-```bash
-agent-browser network requests              # View requests
-agent-browser network route "**/api" --abort  # Block requests
-agent-browser set offline on                # Offline mode
-```
-
-### Debug Mode
-```bash
-agent-browser open example.com --headed     # Show browser window
-agent-browser console                       # View console logs
-agent-browser errors                        # View page errors
-agent-browser highlight @e1                 # Highlight element
-```
-
-## Integration Examples
-
-### With Workflow Skill
-```bash
-# In /workflow execution phase
-agent-browser open $TEST_URL
-agent-browser snapshot -i
-# AI analyzes snapshot, identifies test targets
-agent-browser click @e2
-agent-browser wait --text "Success"
-```
-
-### With TDD Skill
-```bash
-# E2E test execution
-agent-browser open localhost:3000
-agent-browser snapshot -i
-agent-browser fill @e1 "test input"
-agent-browser click @e2
-agent-browser get text @e3  # Verify output
-```
-
-## Best Practices
-
-1. **Always use `-i` flag** for snapshots (interactive elements only)
-2. **Use refs (@e1)** instead of CSS selectors
-3. **Re-snapshot after page changes** to get fresh refs
-4. **Use `--json` flag** for programmatic parsing
-5. **Combine with wait** for dynamic content
-
-## Troubleshooting
-
-```bash
-# Browser not installed
-agent-browser install
-
-# Linux dependencies
-agent-browser install --with-deps
-
-# Debug issues
-agent-browser open url --headed --debug
-```
-
-## Command Reference
-
-| Command | Description |
-|---------|-------------|
-| `open <url>` | Navigate to URL |
-| `snapshot [-i] [-c]` | Get accessibility tree |
-| `click @ref` | Click element |
-| `fill @ref "text"` | Fill input |
-| `get text @ref` | Get text content |
-| `screenshot [path]` | Take screenshot |
-| `wait <ref\|ms\|--text>` | Wait for condition |
-| `close` | Close browser |
-
----
-
-**Zero-Config. AI-Native. Just Works.**

package/templates/common/skills/en/brainstorming.md

@@ -1,64 +0,0 @@
----
-name: brainstorming
-description: Interactive design exploration and refinement
-version: 1.0.0
-author: CCJK
-category: planning
-triggers:
-  - /brainstorm
-  - /bs
-  - /explore
-use_when:
-  - "User wants to explore ideas"
-  - "Starting new feature design"
-  - "User mentions brainstorming or exploring options"
-  - "Need to evaluate different approaches"
-auto_activate: true
-priority: 8
-difficulty: beginner
-tags:
-  - design
-  - exploration
-  - ideation
----
-
-# Brainstorming
-
-## Context
-$ARGUMENTS
-
-## Instructions
-
-You are in brainstorming mode. Your goal is to help explore and refine ideas through interactive discussion.
-
-### Process
-
-1. **Understand the Goal**
-   - What problem are we solving?
-   - What constraints exist?
-   - What does success look like?
-
-2. **Generate Options**
-   - Propose 3-5 different approaches
-   - Consider trade-offs for each
-   - Think about edge cases
-
-3. **Evaluate Together**
-   - Discuss pros and cons
-   - Ask clarifying questions
-   - Refine based on feedback
-
-4. **Converge on Direction**
-   - Summarize the chosen approach
-   - Document key decisions
-   - Identify next steps
-
-### Output Format
-
-Present options as:
-- **Option A**: [Name] - [Brief description]
-  - Pros: ...
-  - Cons: ...
-  - Best for: ...
-
-Ask questions to clarify requirements before finalizing.