09-VS-Code_VSCode集成
VS Code 集成
本实验内容
本实验为您提供全面的指导，帮助您将 MCP 服务器与 VS Code 集成，以通过 AI 聊天实现自然语言查询。您将学习如何配置 VS Code 以优化 MCP 的使用、调试服务器连接，并充分利用 AI 辅助的数据库交互功能。
概述
VS Code 的 MCP 集成彻底改变了开发者通过自然语言与数据库和 API 交互的方式。通过将您的零售 MCP 服务器连接到 VS Code Chat，您可以使用对话式 AI 智能查询销售数据、产品目录和业务分析。
这种集成允许开发者提出诸如“显示本月的畅销产品”或“查找 90 天内未购买的客户”之类的问题，并获得结构化数据响应，而无需编写 SQL 查询。
学习目标
完成本实验后，您将能够：
- 配置 VS Code MCP 设置以连接您的零售服务器
- 集成 MCP 服务器与 VS Code AI Chat 功能
- 调试 MCP 服务器连接并解决问题
- 优化 自然语言查询模式以获得更好的结果
- 自定义 VS Code 工作区以支持 MCP 开发
- 部署 多服务器配置以应对复杂场景
VS Code MCP 配置
初始设置与安装
// .vscode/settings.json { "mcp.servers": { "retail-mcp-server": { "command": "python", "args": [ "-m", "mcp_server.main" ], "env": { "POSTGRES_HOST": "localhost", "POSTGRES_PORT": "5432", "POSTGRES_DB": "retail_db", "POSTGRES_USER": "mcp_user", "POSTGRES_PASSWORD": "${env:POSTGRES_PASSWORD}", "PROJECT_ENDPOINT": "${env:PROJECT_ENDPOINT}", "AZURE_CLIENT_ID": "${env:AZURE_CLIENT_ID}", "AZURE_CLIENT_SECRET": "${env:AZURE_CLIENT_SECRET}", "AZURE_TENANT_ID": "${env:AZURE_TENANT_ID}", "LOG_LEVEL": "INFO", "MCP_SERVER_DEBUG": "false" }, "cwd": "${workspaceFolder}", "initializationOptions": { "store_id": "seattle", "enable_semantic_search": true, "enable_analytics": true, "cache_embeddings": true } } }, "mcp.serverTimeout": 30000, "mcp.enableLogging": true, "mcp.logLevel": "info" }
环境配置
# .env file for development POSTGRES_HOST=localhost POSTGRES_PORT=5432 POSTGRES_DB=retail_db POSTGRES_USER=mcp_user POSTGRES_PASSWORD=your_secure_password # Azure Configuration PROJECT_ENDPOINT=https://your-project.openai.azure.com AZURE_CLIENT_ID=your-client-id AZURE_CLIENT_SECRET=your-client-secret AZURE_TENANT_ID=your-tenant-id # Optional: Azure Key Vault AZURE_KEY_VAULT_URL=https://your-keyvault.vault.azure.net/ # Server Configuration MCP_SERVER_PORT=8000 MCP_SERVER_HOST=127.0.0.1 LOG_LEVEL=INFO
工作区配置
// .vscode/launch.json { "version": "0.2.0", "configurations": [ { "name": "Debug MCP Server", "type": "python", "request": "launch", "module": "mcp_server.main", "console": "integratedTerminal", "envFile": "${workspaceFolder}/.env", "env": { "MCP_SERVER_DEBUG": "true", "LOG_LEVEL": "DEBUG" }, "args": [], "justMyCode": false, "stopOnEntry": false }, { "name": "Test MCP Server", "type": "python", "request": "launch", "module": "pytest", "console": "integratedTerminal", "envFile": "${workspaceFolder}/.env.test", "args": [ "tests/", "-v", "--tb=short" ] } ] }
任务配置
// .vscode/tasks.json { "version": "2.0.0", "tasks": [ { "label": "Start MCP Server", "type": "shell", "command": "python", "args": [ "-m", "mcp_server.main" ], "group": "build", "presentation": { "echo": true, "reveal": "always", "focus": false, "panel": "new" }, "options": { "env": { "PYTHONPATH": "${workspaceFolder}" } }, "isBackground": true, "problemMatcher": { "pattern": { "regexp": "^(.*):(\\d+):(\\d+):\\s+(warning|error):\\s+(.*)$", "file": 1, "line": 2, "column": 3, "severity": 4, "message": 5 }, "background": { "activeOnStart": true, "beginsPattern": "^.*Starting MCP server.*$", "endsPattern": "^.*MCP server ready.*$" } } }, { "label": "Run Tests", "type": "shell", "command": "python", "args": [ "-m", "pytest", "tests/", "-v" ], "group": "test", "presentation": { "echo": true, "reveal": "always", "focus": false, "panel": "shared" } }, { "label": "Generate Sample Data", "type": "shell", "command": "python", "args": [ "scripts/generate_sample_data.py" ], "group": "build", "presentation": { "echo": true, "reveal": "always", "focus": false, "panel": "shared" } }, { "label": "Create Database Schema", "type": "shell", "command": "psql", "args": [ "-h", "${env:POSTGRES_HOST}", "-p", "${env:POSTGRES_PORT}", "-U", "${env:POSTGRES_USER}", "-d", "${env:POSTGRES_DB}", "-f", "scripts/create_schema.sql" ], "group": "build" } ] }
AI 聊天集成
自然语言查询模式
// Example query patterns for VS Code Chat interface QueryPattern { intent: string; examples: string[]; expectedTools: string[]; } const retailQueryPatterns: QueryPattern[] = [ { intent: "sales_analysis", examples: [ "Show me daily sales for the last 30 days", "What are our top selling products this month?", "Which customers have spent the most this quarter?", "Compare sales performance between stores" ], expectedTools: ["execute_sales_query"] }, { intent: "product_search", examples: [ "Find running shoes for women", "Show me electronics under $500", "What laptops do we have in stock?", "Search for wireless headphones" ], expectedTools: ["semantic_search_products", "hybrid_product_search"] }, { intent: "inventory_management", examples: [ "Which products are low on stock?", "Show me products that need reordering", "What's our current inventory value?", "Find products with zero stock" ], expectedTools: ["execute_sales_query"] }, { intent: "customer_analysis", examples: [ "Show me customers who haven't purchased in 90 days", "What's the average customer lifetime value?", "Which customers are in the gold tier?", "Find customers with returns" ], expectedTools: ["execute_sales_query"] }, { intent: "business_intelligence", examples: [ "Generate a business summary for this month", "Show me seasonal trends", "What are our best performing categories?", "Create a sales forecast" ], expectedTools: ["generate_business_insights"] }, { intent: "recommendations", examples: [ "Recommend products similar to product X", "What should we recommend to customer Y?", "Show me trending products", "Find cross-sell opportunities" ], expectedTools: ["get_product_recommendations"] } ];
聊天集成示例
<!-- Examples of VS Code Chat interactions --> ## Sales Analysis Queries **User**: Show me the top 10 selling products in the Seattle store for the last month **Expected Response**: - Tool: execute_sales_query - Parameters: query_type="top_products", store_id="seattle", start_date="2025-08-29", end_date="2025-09-29", limit=10 - Result: Formatted table with product names, quantities sold, revenue, and performance metrics **User**: What was our daily revenue trend last week? **Expected Response**: - Tool: execute_sales_query - Parameters: query_type="daily_sales", store_id="seattle", start_date="2025-09-22", end_date="2025-09-29" - Result: Chart-ready data with daily revenue figures and growth percentages ## Product Search Queries **User**: Find comfortable running shoes for outdoor activities **Expected Response**: - Tool: semantic_search_products - Parameters: query="comfortable running shoes outdoor activities", store_id="seattle", similarity_threshold=0.7 - Result: Ranked list of relevant products with similarity scores and detailed information **User**: Search for laptops under $1500 with good reviews **Expected Response**: - Tool: hybrid_product_search - Parameters: query="laptops under $1500 good reviews", store_id="seattle", semantic_weight=0.6, keyword_weight=0.4 - Result: Combined keyword and semantic search results with price and rating filters ## Business Intelligence Queries **User**: Generate a comprehensive business summary for September **Expected Response**: - Tool: generate_business_insights - Parameters: analysis_type="summary", store_id="seattle", days=30 - Result: KPI dashboard with revenue, customer metrics, top categories, and growth trends
聊天响应格式化
# mcp_server/chat/response_formatter.py """ Format MCP tool responses for optimal VS Code Chat display. """ from typing import Dict, Any, List import json from datetime import datetime class ChatResponseFormatter: """Format tool responses for VS Code Chat consumption.""" @staticmethod def format_sales_data(data: List[Dict[str, Any]], query_type: str) -> str: """Format sales data for chat display.""" if not data: return "No sales data found for the specified criteria." if query_type == "daily_sales": return ChatResponseFormatter._format_daily_sales(data) elif query_type == "top_products": return ChatResponseFormatter._format_top_products(data) elif query_type == "customer_analysis": return ChatResponseFormatter._format_customer_analysis(data) else: return ChatResponseFormatter._format_generic_table(data) @staticmethod def _format_daily_sales(data: List[Dict[str, Any]]) -> str: """Format daily sales data.""" response = "## Daily Sales Summary\n\n" response += "| Date | Revenue | Transactions | Avg Order Value | Customers |\n" response += "|------|---------|-------------|----------------|----------|\n" total_revenue = 0 total_transactions = 0 for day in data: revenue = float(day.get('total_revenue', 0)) transactions = int(day.get('transaction_count', 0)) avg_value = float(day.get('avg_transaction_value', 0)) customers = int(day.get('unique_customers', 0)) total_revenue += revenue total_transactions += transactions response += f"| {day.get('sales_date', 'N/A')} | " response += f"${revenue:,.2f} | " response += f"{transactions:,} | " response += f"${avg_value:.2f} | " response += f"{customers:,} |\n" response += f"\n**Totals**: ${total_revenue:,.2f} revenue, {total_transactions:,} transactions" return response @staticmethod def _format_top_products(data: List[Dict[str, Any]]) -> str: """Format top products data.""" response = "## Top Selling Products\n\n" response += "| Rank | Product | Brand | Revenue | Qty Sold | Avg Price |\n" response += "|------|---------|-------|---------|----------|----------|\n" for i, product in enumerate(data, 1): response += f"| {i} | " response += f"{product.get('product_name', 'N/A')} | " response += f"{product.get('brand', 'N/A')} | " response += f"${float(product.get('total_revenue', 0)):,.2f} | " response += f"{int(product.get('total_quantity_sold', 0)):,} | " response += f"${float(product.get('avg_price', 0)):.2f} |\n" return response @staticmethod def format_search_results(data: List[Dict[str, Any]], search_type: str) -> str: """Format product search results.""" if not data: return "No products found matching your search criteria." response = f"## Product Search Results ({search_type})\n\n" for i, product in enumerate(data, 1): response += f"### {i}. {product.get('product_name', 'Unknown Product')}\n" response += f"**Brand**: {product.get('brand', 'N/A')}\n" response += f"**Price**: ${float(product.get('price', 0)):.2f}\n" response += f"**Stock**: {int(product.get('current_stock', 0))} units\n" if 'similarity_score' in product: score = float(product['similarity_score']) response += f"**Relevance**: {score:.1%}\n" if 'rating_average' in product and product['rating_average']: rating = float(product['rating_average']) count = int(product.get('rating_count', 0)) response += f"**Rating**: {rating:.1f}/5.0 ({count:,} reviews)\n" if product.get('product_description'): desc = product['product_description'] if len(desc) > 150: desc = desc[:150] + "..." response += f"**Description**: {desc}\n" response += "\n---\n\n" return response @staticmethod def format_business_insights(data: Dict[str, Any]) -> str: """Format business intelligence data.""" response = "## Business Intelligence Summary\n\n" # Key metrics response += "### Key Performance Indicators\n\n" response += f"- **Total Revenue**: ${float(data.get('total_revenue', 0)):,.2f}\n" response += f"- **Total Transactions**: {int(data.get('total_transactions', 0)):,}\n" response += f"- **Unique Customers**: {int(data.get('unique_customers', 0)):,}\n" response += f"- **Average Order Value**: ${float(data.get('avg_transaction_value', 0)):.2f}\n" response += f"- **Products Sold**: {int(data.get('products_sold', 0)):,} items\n\n" # Performance indicators if 'insights' in data and 'performance_indicators' in data['insights']: pi = data['insights']['performance_indicators'] response += "### Performance Indicators\n\n" response += f"- **Transactions per Day**: {float(pi.get('transactions_per_day', 0)):.1f}\n" response += f"- **Revenue per Customer**: ${float(pi.get('revenue_per_customer', 0)):,.2f}\n" response += f"- **Items per Transaction**: {float(pi.get('items_per_transaction', 0)):.1f}\n\n" # Top category if data.get('top_category'): response += f"### Top Performing Category\n\n" response += f"**{data['top_category']}** - ${float(data.get('top_category_revenue', 0)):,.2f} revenue\n\n" return response @staticmethod def format_error_response(error: str, tool_name: str) -> str: """Format error responses for chat.""" response = f"## ❌ Error in {tool_name}\n\n" response += f"I encountered an issue while processing your request:\n\n" response += f"**Error**: {error}\n\n" response += "Please try:\n" response += "- Checking your query parameters\n" response += "- Verifying store access permissions\n" response += "- Simplifying your request\n" response += "- Contacting support if the issue persists\n" return response
调试与故障排除
VS Code 调试配置
# mcp_server/debug/vscode_debug.py """ VS Code specific debugging utilities for MCP server. """ import logging import json from typing import Dict, Any from datetime import datetime class VSCodeDebugLogger: """Enhanced logging for VS Code debugging.""" def __init__(self): self.logger = logging.getLogger("mcp_vscode_debug") self.setup_vscode_logging() def setup_vscode_logging(self): """Configure logging for VS Code debugging.""" # Create VS Code specific formatter formatter = logging.Formatter( '[%(asctime)s] [%(name)s] [%(levelname)s] %(message)s' ) # Console handler for VS Code terminal console_handler = logging.StreamHandler() console_handler.setFormatter(formatter) console_handler.setLevel(logging.DEBUG) self.logger.addHandler(console_handler) self.logger.setLevel(logging.DEBUG) def log_mcp_request(self, method: str, params: Dict[str, Any]): """Log MCP requests for debugging.""" self.logger.info(f"MCP Request: {method}") self.logger.debug(f"Parameters: {json.dumps(params, indent=2)}") def log_tool_execution(self, tool_name: str, result: Dict[str, Any]): """Log tool execution results.""" success = result.get('success', False) level = logging.INFO if success else logging.ERROR self.logger.log(level, f"Tool '{tool_name}' - {'Success' if success else 'Failed'}") if not success and result.get('error'): self.logger.error(f"Error: {result['error']}") if result.get('data'): data_summary = self._summarize_data(result['data']) self.logger.debug(f"Result summary: {data_summary}") def _summarize_data(self, data: Any) -> str: """Create a summary of result data.""" if isinstance(data, list): return f"List with {len(data)} items" elif isinstance(data, dict): return f"Dict with keys: {list(data.keys())}" else: return f"Data type: {type(data).__name__}" # Global debug logger vscode_debug_logger = VSCodeDebugLogger()
连接故障排除
# scripts/debug_mcp_connection.py """ Debug script for troubleshooting MCP server connections in VS Code. """ import asyncio import asyncpg import os import sys from typing import Dict, Any async def test_database_connection() -> Dict[str, Any]: """Test database connectivity.""" try: # Get connection parameters from environment connection_params = { 'host': os.getenv('POSTGRES_HOST', 'localhost'), 'port': int(os.getenv('POSTGRES_PORT', '5432')), 'database': os.getenv('POSTGRES_DB', 'retail_db'), 'user': os.getenv('POSTGRES_USER', 'mcp_user'), 'password': os.getenv('POSTGRES_PASSWORD', '') } print(f"Testing connection to {connection_params['host']}:{connection_params['port']}") # Test connection conn = await asyncpg.connect(**connection_params) # Test basic query result = await conn.fetchval("SELECT version()") # Test schema access tables = await conn.fetch(""" SELECT table_name FROM information_schema.tables WHERE table_schema = 'retail' """) await conn.close() return { 'success': True, 'database_version': result, 'retail_tables': len(tables), 'table_names': [table['table_name'] for table in tables] } except Exception as e: return { 'success': False, 'error': str(e), 'connection_params': {k: v for k, v in connection_params.items() if k != 'password'} } async def test_azure_openai_connection() -> Dict[str, Any]: """Test Azure OpenAI connectivity.""" try: from azure.identity import DefaultAzureCredential from azure.ai.projects import AIProjectClient project_endpoint = os.getenv('PROJECT_ENDPOINT') if not project_endpoint: return { 'success': False, 'error': 'PROJECT_ENDPOINT not configured' } print(f"Testing Azure OpenAI connection to {project_endpoint}") credential = DefaultAzureCredential() client = AIProjectClient( endpoint=project_endpoint, credential=credential ) # Test embedding generation response = await client.embeddings.create( model="text-embedding-3-small", input="test connection" ) embedding = response.data[0].embedding return { 'success': True, 'project_endpoint': project_endpoint, 'embedding_dimension': len(embedding), 'model': 'text-embedding-3-small' } except Exception as e: return { 'success': False, 'error': str(e), 'project_endpoint': os.getenv('PROJECT_ENDPOINT', 'Not configured') } async def test_mcp_tools() -> Dict[str, Any]: """Test MCP tool availability.""" try: # Import MCP server components sys.path.append(os.path.dirname(os.path.dirname(__file__))) from mcp_server.server import MCPServer from mcp_server.database import DatabaseProvider from mcp_server.config import Config # Create test configuration config = Config() db_provider = DatabaseProvider(config.database.connection_string) # Initialize server server = MCPServer(config, db_provider) await server.initialize() # Get available tools tools = server.get_available_tools() # Test a simple tool test_result = await server.execute_tool( 'get_current_utc_date', {'format': 'iso'} ) await server.cleanup() return { 'success': True, 'available_tools': [tool.name for tool in tools], 'tool_count': len(tools), 'test_tool_result': test_result.get('success', False) } except Exception as e: return { 'success': False, 'error': str(e) } async def main(): """Run comprehensive connection tests.""" print(" MCP Server Connection Diagnostics") print("=" * 50) # Test database connection print("\n Testing Database Connection...") db_result = await test_database_connection() if db_result['success']: print("✅ Database connection successful") print(f" Database version: {db_result['database_version']}") print(f" Retail tables found: {db_result['retail_tables']}") print(f" Table names: {', '.join(db_result['table_names'])}") else: print("❌ Database connection failed") print(f" Error: {db_result['error']}") # Test Azure OpenAI connection print("\n Testing Azure OpenAI Connection...") azure_result = await test_azure_openai_connection() if azure_result['success']: print("✅ Azure OpenAI connection successful") print(f" Endpoint: {azure_result['project_endpoint']}") print(f" Embedding dimension: {azure_result['embedding_dimension']}") else: print("❌ Azure OpenAI connection failed") print(f" Error: {azure_result['error']}") # Test MCP tools print("\n️ Testing MCP Tools...") tools_result = await test_mcp_tools() if tools_result['success']: print("✅ MCP tools loaded successfully") print(f" Available tools: {tools_result['tool_count']}") print(f" Tool names: {', '.join(tools_result['available_tools'])}") print(f" Test execution: {'✅' if tools_result['test_tool_result'] else '❌'}") else: print("❌ MCP tools loading failed") print(f" Error: {tools_result['error']}") # Overall status print("\n Overall Status") print("=" * 50) all_success = all([ db_result['success'], azure_result['success'], tools_result['success'] ]) if all_success: print(" All systems ready! MCP server should work correctly in VS Code.") else: print("⚠️ Some issues detected. Please resolve the errors above.") print("\n Troubleshooting tips:") print(" - Check environment variables in .env file") print(" - Verify database is running and accessible") print(" - Confirm Azure credentials are configured") print(" - Review VS Code MCP server configuration") if __name__ == "__main__": asyncio.run(main())
高级配置
多服务器设置
// .vscode/settings.json - Multiple MCP servers { "mcp.servers": { "retail-seattle": { "command": "python", "args": ["-m", "mcp_server.main"], "env": { "POSTGRES_HOST": "localhost", "POSTGRES_DB": "retail_db", "POSTGRES_USER": "mcp_user", "POSTGRES_PASSWORD": "${env:POSTGRES_PASSWORD}", "PROJECT_ENDPOINT": "${env:PROJECT_ENDPOINT}", "DEFAULT_STORE_ID": "seattle" }, "initializationOptions": { "store_id": "seattle", "server_name": "Seattle Store" } }, "retail-redmond": { "command": "python", "args": ["-m", "mcp_server.main"], "env": { "POSTGRES_HOST": "localhost", "POSTGRES_DB": "retail_db", "POSTGRES_USER": "mcp_user", "POSTGRES_PASSWORD": "${env:POSTGRES_PASSWORD}", "PROJECT_ENDPOINT": "${env:PROJECT_ENDPOINT}", "DEFAULT_STORE_ID": "redmond" }, "initializationOptions": { "store_id": "redmond", "server_name": "Redmond Store" } }, "retail-analytics": { "command": "python", "args": ["-m", "mcp_server.analytics_main"], "env": { "POSTGRES_HOST": "localhost", "POSTGRES_DB": "retail_db", "POSTGRES_USER": "analytics_user", "POSTGRES_PASSWORD": "${env:ANALYTICS_PASSWORD}", "PROJECT_ENDPOINT": "${env:PROJECT_ENDPOINT}" }, "initializationOptions": { "mode": "analytics", "cross_store_access": true } } } }
自定义 VS Code 扩展
// src/extension.ts - Custom MCP retail extension import * as vscode from 'vscode'; export function activate(context: vscode.ExtensionContext) { // Register MCP retail commands const disposable = vscode.commands.registerCommand( 'mcp-retail.quickQuery', async () => { const quickPick = vscode.window.createQuickPick(); quickPick.items = [ { label: ' Daily Sales', description: 'Show daily sales for the last 30 days' }, { label: ' Top Products', description: 'Show top selling products this month' }, { label: ' Customer Analysis', description: 'Analyze customer behavior and trends' }, { label: ' Product Search', description: 'Search for products using natural language' }, { label: ' Business Insights', description: 'Generate comprehensive business summary' } ]; quickPick.onDidChangeSelection(selection => { if (selection[0]) { executeQuickQuery(selection[0].label); } }); quickPick.onDidHide(() => quickPick.dispose()); quickPick.show(); } ); context.subscriptions.push(disposable); // Register store switcher const storeSwitcher = vscode.commands.registerCommand( 'mcp-retail.switchStore', async () => { const stores = ['seattle', 'redmond', 'bellevue', 'online']; const selected = await vscode.window.showQuickPick(stores, { placeHolder: 'Select store for queries' }); if (selected) { // Update configuration const config = vscode.workspace.getConfiguration('mcp'); await config.update('defaultStore', selected, true); vscode.window.showInformationMessage( `Switched to ${selected.charAt(0).toUpperCase() + selected.slice(1)} store` ); } } ); context.subscriptions.push(storeSwitcher); } async function executeQuickQuery(queryType: string) { // Execute predefined queries in VS Code Chat const chatCommands = { ' Daily Sales': '@retail Show me daily sales for the last 30 days', ' Top Products': '@retail What are the top 10 selling products this month?', ' Customer Analysis': '@retail Show me customer analysis for active customers', ' Product Search': '@retail Find products matching "laptop computer"', ' Business Insights': '@retail Generate a business summary for this month' }; const command = chatCommands[queryType]; if (command) { await vscode.commands.executeCommand('workbench.action.chat.open'); await vscode.commands.executeCommand('workbench.action.chat.insert', command); } } export function deactivate() {}
扩展包配置
// package.json for VS Code extension { "name": "mcp-retail-assistant", "displayName": "MCP Retail Assistant", "description": "AI-powered retail data analysis through MCP", "version": "1.0.0", "engines": { "vscode": "^1.74.0" }, "categories": [ "Other", "Data Science", "Machine Learning" ], "activationEvents": [ "onCommand:mcp-retail.quickQuery", "onCommand:mcp-retail.switchStore" ], "main": "./out/extension.js", "contributes": { "commands": [ { "command": "mcp-retail.quickQuery", "title": "Quick Retail Query", "category": "MCP Retail" }, { "command": "mcp-retail.switchStore", "title": "Switch Store", "category": "MCP Retail" } ], "keybindings": [ { "command": "mcp-retail.quickQuery", "key": "ctrl+shift+r", "mac": "cmd+shift+r" } ], "configuration": { "title": "MCP Retail", "properties": { "mcp-retail.defaultStore": { "type": "string", "default": "seattle", "enum": ["seattle", "redmond", "bellevue", "online"], "description": "Default store for retail queries" }, "mcp-retail.enableAnalytics": { "type": "boolean", "default": true, "description": "Enable advanced analytics features" } } } }, "scripts": { "vscode:prepublish": "npm run compile", "compile": "tsc -p ./", "watch": "tsc -watch -p ./" }, "devDependencies": { "@types/vscode": "^1.74.0", "@types/node": "16.x", "typescript": "^4.9.4" } }
关键收获
完成本实验后，您将掌握：
✅ VS Code MCP 配置：完成 MCP 集成的最佳设置
✅ AI 聊天集成：在 VS Code 中实现自然语言查询功能
✅ 调试工具：全面的故障排除和连接诊断能力
✅ 多服务器设置：配置多个 MCP 服务器实例
✅ 自定义扩展：通过零售特定功能增强 VS Code 体验
✅ 生产环境准备：企业级 VS Code 开发环境
下一步
继续学习 实验 10：部署策略，以了解：
- 将 MCP 服务器部署到生产环境
- 配置云基础设施以实现可扩展性
- 实现 CI/CD 流水线以自动化部署
- 监控生产环境 MCP 服务器性能
其他资源
VS Code 开发
- VS Code Extension API - 官方扩展开发指南
- VS Code MCP 文档 - MCP 集成文档
- VS Code 的 TypeScript - 在 VS Code 中进行 TypeScript 开发
MCP 协议
- 模型上下文协议规范 - 官方 MCP 规范
- MCP 最佳实践 - 实现最佳实践
- FastMCP 框架 - Python MCP 实现
开发工具
- VS Code 中的 Python - Python 开发设置
- VS Code 调试 - 高级调试技术
- VS Code 任务 - 任务自动化与配置
上一节: 实验 08：测试与调试
下一节: 实验 10：部署策略
免责声明：
本文档使用AI翻译服务 Co-op Translator 进行翻译。尽管我们努力确保翻译的准确性，但请注意，自动翻译可能包含错误或不准确之处。原始语言的文档应被视为权威来源。对于关键信息，建议使用专业人工翻译。我们不对因使用此翻译而产生的任何误解或误读承担责任。
评论区
(0)
U