@mison/ag-kit-cn 2.0.1

package/.agent/skills/lint-and-validate/scripts/lint_runner.py

@@ -0,0 +1,184 @@
+#!/usr/bin/env python3
+"""
+Lint Runner - Unified linting and type checking
+Runs appropriate linters based on project type.
+
+Usage:
+    python lint_runner.py <project_path>
+
+Supports:
+    - Node.js: npm run lint, npx tsc --noEmit
+    - Python: ruff check, mypy
+"""
+
+import subprocess
+import sys
+import json
+import platform
+import shutil
+from pathlib import Path
+from datetime import datetime
+
+# Fix Windows console encoding
+try:
+    sys.stdout.reconfigure(encoding='utf-8', errors='replace')
+except:
+    pass
+
+
+def detect_project_type(project_path: Path) -> dict:
+    """Detect project type and available linters."""
+    result = {
+        "type": "unknown",
+        "linters": []
+    }
+    
+    # Node.js project
+    package_json = project_path / "package.json"
+    if package_json.exists():
+        result["type"] = "node"
+        try:
+            pkg = json.loads(package_json.read_text(encoding='utf-8'))
+            scripts = pkg.get("scripts", {})
+            deps = {**pkg.get("dependencies", {}), **pkg.get("devDependencies", {})}
+            
+            # Check for lint script
+            if "lint" in scripts:
+                result["linters"].append({"name": "npm lint", "cmd": ["npm", "run", "lint"]})
+            elif "eslint" in deps:
+                result["linters"].append({"name": "eslint", "cmd": ["npx", "eslint", "."]})
+            
+            # Check for TypeScript
+            if "typescript" in deps or (project_path / "tsconfig.json").exists():
+                result["linters"].append({"name": "tsc", "cmd": ["npx", "tsc", "--noEmit"]})
+                
+        except:
+            pass
+    
+    # Python project
+    if (project_path / "pyproject.toml").exists() or (project_path / "requirements.txt").exists():
+        result["type"] = "python"
+        
+        # Check for ruff
+        result["linters"].append({"name": "ruff", "cmd": ["ruff", "check", "."]})
+        
+        # Check for mypy
+        if (project_path / "mypy.ini").exists() or (project_path / "pyproject.toml").exists():
+            result["linters"].append({"name": "mypy", "cmd": ["mypy", "."]})
+    
+    return result
+
+
+def run_linter(linter: dict, cwd: Path) -> dict:
+    """Run a single linter and return results."""
+    result = {
+        "name": linter["name"],
+        "passed": False,
+        "output": "",
+        "error": ""
+    }
+    
+    try:
+        cmd = linter["cmd"]
+        
+        # Windows compatibility for npm/npx
+        if platform.system() == "Windows":
+            if cmd[0] in ["npm", "npx"]:
+                # Force .cmd extension on Windows
+                if not cmd[0].lower().endswith(".cmd"):
+                    cmd[0] = f"{cmd[0]}.cmd"
+        
+        proc = subprocess.run(
+            cmd,
+            cwd=str(cwd),
+            capture_output=True,
+            text=True,
+            encoding='utf-8',
+            errors='replace',
+            timeout=120,
+            shell=platform.system() == "Windows" # Shell=True often helps with path resolution on Windows
+        )
+        
+        result["output"] = proc.stdout[:2000] if proc.stdout else ""
+        result["error"] = proc.stderr[:500] if proc.stderr else ""
+        result["passed"] = proc.returncode == 0
+        
+    except FileNotFoundError:
+        result["error"] = f"Command not found: {linter['cmd'][0]}"
+    except subprocess.TimeoutExpired:
+        result["error"] = "Timeout after 120s"
+    except Exception as e:
+        result["error"] = str(e)
+    
+    return result
+
+
+def main():
+    project_path = Path(sys.argv[1] if len(sys.argv) > 1 else ".").resolve()
+    
+    print(f"\n{'='*60}")
+    print(f"[LINT RUNNER] Unified Linting")
+    print(f"{'='*60}")
+    print(f"Project: {project_path}")
+    print(f"Time: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}")
+    
+    # Detect project type
+    project_info = detect_project_type(project_path)
+    print(f"Type: {project_info['type']}")
+    print(f"Linters: {len(project_info['linters'])}")
+    print("-"*60)
+    
+    if not project_info["linters"]:
+        print("No linters found for this project type.")
+        output = {
+            "script": "lint_runner",
+            "project": str(project_path),
+            "type": project_info["type"],
+            "checks": [],
+            "passed": True,
+            "message": "No linters configured"
+        }
+        print(json.dumps(output, indent=2))
+        sys.exit(0)
+    
+    # Run each linter
+    results = []
+    all_passed = True
+    
+    for linter in project_info["linters"]:
+        print(f"\nRunning: {linter['name']}...")
+        result = run_linter(linter, project_path)
+        results.append(result)
+        
+        if result["passed"]:
+            print(f"  [PASS] {linter['name']}")
+        else:
+            print(f"  [FAIL] {linter['name']}")
+            if result["error"]:
+                print(f"  Error: {result['error'][:200]}")
+            all_passed = False
+    
+    # Summary
+    print("\n" + "="*60)
+    print("SUMMARY")
+    print("="*60)
+    
+    for r in results:
+        icon = "[PASS]" if r["passed"] else "[FAIL]"
+        print(f"{icon} {r['name']}")
+    
+    output = {
+        "script": "lint_runner",
+        "project": str(project_path),
+        "type": project_info["type"],
+        "checks": results,
+        "passed": all_passed
+    }
+    
+    print("\n" + json.dumps(output, indent=2))
+    
+    sys.exit(0 if all_passed else 1)
+
+
+if __name__ == "__main__":
+    main()

package/.agent/skills/lint-and-validate/scripts/type_coverage.py

@@ -0,0 +1,173 @@
+#!/usr/bin/env python3
+"""
+Type Coverage Checker - Measures TypeScript/Python type coverage.
+Identifies untyped functions, any usage, and type safety issues.
+"""
+import sys
+import re
+import subprocess
+from pathlib import Path
+
+# Fix Windows console encoding for Unicode output
+try:
+    sys.stdout.reconfigure(encoding='utf-8', errors='replace')
+    sys.stderr.reconfigure(encoding='utf-8', errors='replace')
+except AttributeError:
+    pass  # Python < 3.7
+
+def check_typescript_coverage(project_path: Path) -> dict:
+    """Check TypeScript type coverage."""
+    issues = []
+    passed = []
+    stats = {'any_count': 0, 'untyped_functions': 0, 'total_functions': 0}
+    
+    ts_files = list(project_path.rglob("*.ts")) + list(project_path.rglob("*.tsx"))
+    ts_files = [f for f in ts_files if 'node_modules' not in str(f) and '.d.ts' not in str(f)]
+    
+    if not ts_files:
+        return {'type': 'typescript', 'files': 0, 'passed': [], 'issues': ["[!] No TypeScript files found"], 'stats': stats}
+    
+    for file_path in ts_files[:30]:  # Limit
+        try:
+            content = file_path.read_text(encoding='utf-8', errors='ignore')
+            
+            # Count 'any' usage
+            any_matches = re.findall(r':\s*any\b', content)
+            stats['any_count'] += len(any_matches)
+            
+            # Find functions without return types
+            # function name(params) { - no return type
+            untyped = re.findall(r'function\s+\w+\s*\([^)]*\)\s*{', content)
+            # Arrow functions without types: const fn = (x) => or (x) =>
+            untyped += re.findall(r'=\s*\([^:)]*\)\s*=>', content)
+            stats['untyped_functions'] += len(untyped)
+            
+            # Count typed functions
+            typed = re.findall(r'function\s+\w+\s*\([^)]*\)\s*:\s*\w+', content)
+            typed += re.findall(r':\s*\([^)]*\)\s*=>\s*\w+', content)
+            stats['total_functions'] += len(typed) + len(untyped)
+            
+        except Exception:
+            continue
+    
+    # Analyze results
+    if stats['any_count'] == 0:
+        passed.append("[OK] No 'any' types found")
+    elif stats['any_count'] <= 5:
+        issues.append(f"[!] {stats['any_count']} 'any' types found (acceptable)")
+    else:
+        issues.append(f"[X] {stats['any_count']} 'any' types found (too many)")
+    
+    if stats['total_functions'] > 0:
+        typed_ratio = (stats['total_functions'] - stats['untyped_functions']) / stats['total_functions'] * 100
+        if typed_ratio >= 80:
+            passed.append(f"[OK] Type coverage: {typed_ratio:.0f}%")
+        elif typed_ratio >= 50:
+            issues.append(f"[!] Type coverage: {typed_ratio:.0f}% (improve)")
+        else:
+            issues.append(f"[X] Type coverage: {typed_ratio:.0f}% (too low)")
+    
+    passed.append(f"[OK] Analyzed {len(ts_files)} TypeScript files")
+    
+    return {'type': 'typescript', 'files': len(ts_files), 'passed': passed, 'issues': issues, 'stats': stats}
+
+def check_python_coverage(project_path: Path) -> dict:
+    """Check Python type hints coverage."""
+    issues = []
+    passed = []
+    stats = {'untyped_functions': 0, 'typed_functions': 0, 'any_count': 0}
+    
+    py_files = list(project_path.rglob("*.py"))
+    py_files = [f for f in py_files if not any(x in str(f) for x in ['venv', '__pycache__', '.git', 'node_modules'])]
+    
+    if not py_files:
+        return {'type': 'python', 'files': 0, 'passed': [], 'issues': ["[!] No Python files found"], 'stats': stats}
+    
+    for file_path in py_files[:30]:  # Limit
+        try:
+            content = file_path.read_text(encoding='utf-8', errors='ignore')
+            
+            # Count Any usage
+            any_matches = re.findall(r':\s*Any\b', content)
+            stats['any_count'] += len(any_matches)
+            
+            # Find functions with type hints
+            typed_funcs = re.findall(r'def\s+\w+\s*\([^)]*:[^)]+\)', content)
+            typed_funcs += re.findall(r'def\s+\w+\s*\([^)]*\)\s*->', content)
+            stats['typed_functions'] += len(typed_funcs)
+            
+            # Find functions without type hints
+            all_funcs = re.findall(r'def\s+\w+\s*\(', content)
+            stats['untyped_functions'] += len(all_funcs) - len(typed_funcs)
+            
+        except Exception:
+            continue
+    
+    total = stats['typed_functions'] + stats['untyped_functions']
+    
+    if total > 0:
+        typed_ratio = stats['typed_functions'] / total * 100
+        if typed_ratio >= 70:
+            passed.append(f"[OK] Type hints coverage: {typed_ratio:.0f}%")
+        elif typed_ratio >= 40:
+            issues.append(f"[!] Type hints coverage: {typed_ratio:.0f}%")
+        else:
+            issues.append(f"[X] Type hints coverage: {typed_ratio:.0f}% (add type hints)")
+    
+    if stats['any_count'] == 0:
+        passed.append("[OK] No 'Any' types found")
+    elif stats['any_count'] <= 3:
+        issues.append(f"[!] {stats['any_count']} 'Any' types found")
+    else:
+        issues.append(f"[X] {stats['any_count']} 'Any' types found")
+    
+    passed.append(f"[OK] Analyzed {len(py_files)} Python files")
+    
+    return {'type': 'python', 'files': len(py_files), 'passed': passed, 'issues': issues, 'stats': stats}
+
+def main():
+    target = sys.argv[1] if len(sys.argv) > 1 else "."
+    project_path = Path(target)
+    
+    print("\n" + "=" * 60)
+    print("  TYPE COVERAGE CHECKER")
+    print("=" * 60 + "\n")
+    
+    results = []
+    
+    # Check TypeScript
+    ts_result = check_typescript_coverage(project_path)
+    if ts_result['files'] > 0:
+        results.append(ts_result)
+    
+    # Check Python
+    py_result = check_python_coverage(project_path)
+    if py_result['files'] > 0:
+        results.append(py_result)
+    
+    if not results:
+        print("[!] No TypeScript or Python files found.")
+        sys.exit(0)
+    
+    # Print results
+    critical_issues = 0
+    for result in results:
+        print(f"\n[{result['type'].upper()}]")
+        print("-" * 40)
+        for item in result['passed']:
+            print(f"  {item}")
+        for item in result['issues']:
+            print(f"  {item}")
+            if item.startswith("[X]"):
+                critical_issues += 1
+    
+    print("\n" + "=" * 60)
+    if critical_issues == 0:
+        print("[OK] TYPE COVERAGE: ACCEPTABLE")
+        sys.exit(0)
+    else:
+        print(f"[X] TYPE COVERAGE: {critical_issues} critical issues")
+        sys.exit(1)
+
+if __name__ == "__main__":
+    main()

package/.agent/skills/mcp-builder/SKILL.md

@@ -0,0 +1,176 @@
+---
+name: mcp-builder
+description: MCP（Model Context Protocol）服务器构建原则。工具设计、资源模式与最佳实践。
+allowed-tools: Read, Write, Edit, Glob, Grep
+---
+
+# MCP 构建器
+
+> MCP 服务器构建原则。
+
+---
+
+## 1. MCP 概览
+
+### 什么是 MCP？
+
+Model Context Protocol：用于连接 AI 系统与外部工具、数据源的标准协议。
+
+### 核心概念
+
+| 概念 | 目的 |
+| --- | --- |
+| **Tools** | AI 可调用的函数能力 |
+| **Resources** | AI 可读取的数据资源 |
+| **Prompts** | 预定义提示词模板 |
+
+---
+
+## 2. 服务架构
+
+### 项目结构
+
+```
+my-mcp-server/
+├── src/
+│   └── index.ts      # 主入口
+├── package.json
+└── tsconfig.json
+```
+
+### 传输类型
+
+| 类型 | 用途 |
+| --- | --- |
+| **Stdio** | 本地、CLI 场景 |
+| **SSE** | Web 场景、流式输出 |
+| **WebSocket** | 实时、双向通信 |
+
+---
+
+## 3. 工具设计原则
+
+### 优秀工具设计
+
+| 原则 | 描述 |
+| --- | --- |
+| 名称清晰 | 动作导向（get_weather, create_user） |
+| 单一职责 | 一次只做好一件事 |
+| 输入可校验 | 使用含类型与描述的 schema |
+| 输出结构化 | 响应格式可预测 |
+
+### 输入 Schema 设计
+
+| 字段 | 必填？ |
+| --- | --- |
+| Type | 是（object） |
+| Properties | 定义每个参数 |
+| Required | 列出必填参数 |
+| Description | 人类可读说明 |
+
+---
+
+## 4. 资源模式
+
+### 资源类型
+
+| 类型 | 用途 |
+| --- | --- |
+| Static | 固定数据（配置、文档） |
+| Dynamic | 按请求动态生成 |
+| Template | 含参数的 URI 模板 |
+
+### URI 模式
+
+| 模式 | 示例 |
+| --- | --- |
+| Fixed | `docs://readme` |
+| Parameterized | `users://{userId}` |
+| Collection | `files://project/*` |
+
+---
+
+## 5. 错误处理
+
+### 错误类型
+
+| 情况 | 响应 |
+| --- | --- |
+| 参数无效 | 返回明确的校验错误 |
+| 资源不存在 | 返回清晰的 “not found” |
+| 服务器错误 | 返回通用错误并记录日志 |
+
+### 最佳实践
+
+- 返回结构化错误对象
+- 不暴露内部实现细节
+- 记录可追踪日志用于调试
+- 提供可执行的错误提示
+
+---
+
+## 6. 多模态处理
+
+### 支持类型
+
+| 类型 | 编码 |
+| --- | --- |
+| Text | Plain text |
+| Images | Base64 + MIME type |
+| Files | Base64 + MIME type |
+
+---
+
+## 7. 安全原则
+
+### 输入校验
+
+- 校验所有工具输入
+- 清洗用户提供的数据
+- 限制资源访问范围
+
+### API 密钥
+
+- 使用环境变量存储
+- 禁止日志输出密钥
+- 校验调用权限
+
+---
+
+## 8. 配置
+
+### Claude Desktop 配置
+
+| 字段 | 目的 |
+| --- | --- |
+| command | 要执行的命令 |
+| args | 命令参数 |
+| env | 环境变量 |
+
+---
+
+## 9. 测试
+
+### 测试类别
+
+| 类型 | 关注点 |
+| --- | --- |
+| Unit | 工具逻辑 |
+| Integration | 服务整体 |
+| Contract | Schema 校验 |
+
+---
+
+## 10. 最佳实践检查清单
+
+- [ ] 工具名称清晰且动作导向
+- [ ] 输入 schema 完整且含说明
+- [ ] 输出为结构化 JSON
+- [ ] 覆盖各类错误处理场景
+- [ ] 输入校验到位
+- [ ] 使用环境变量进行配置
+- [ ] 日志可用于排障与调试
+
+---
+
+> **记住：** MCP 工具应保持简单、聚焦且文档完善。AI 依赖描述来正确调用它们。