@performance-agent/mcp-server 1.0.1

package/README.md

@@ -0,0 +1,503 @@
+# Performance Agent MCP Server
+
+A Model Context Protocol (MCP) server that provides 3 core development tools for VS Code and GitHub Copilot.
+
+## 🎯 Features
+
+| Tool | Description |
+|------|-------------|
+| **generateJMXFromSwagger** | Convert Swagger/OpenAPI specs to JMeter performance test scripts |
+| **analyzeCodeComplexity** | Analyze cyclomatic complexity in local git repositories |
+| **profileFunctionPerformance** | Profile function execution time and identify slow functions |
+
+---
+
+## 📦 Installation
+
+### Prerequisites
+- Node.js 18+
+- npm or yarn
+- Git
+- VS Code with GitHub Copilot
+
+### Install Dependencies
+
+```bash
+cd mcp-server
+npm install
+npm run build
+```
+
+### Optional: Enhanced Python Analysis
+
+For more accurate Python complexity analysis:
+
+```bash
+pip install radon
+```
+
+> **Note:** JavaScript/TypeScript analysis works out-of-the-box using `escomplex` (pure npm, no Python needed).
+
+---
+
+## 🔧 Configuration
+
+### 🏠 LOCAL Development Setup
+
+For local development and testing:
+
+#### Step 1: Create/Edit `.vscode/mcp.json`
+
+```json
+{
+  "servers": {
+    "performance-agent": {
+      "command": "node",
+      "args": [
+        "/Users/amanraj/Documents/CodeWorkspace/Performance-Agent/mcp-server/dist/index.js"
+      ],
+      "env": {
+        "PERFORMANCE_AGENT_API_URL": "http://127.0.0.1:8000"
+      }
+    }
+  }
+}
+```
+
+> **Replace** `/absolute/path/to/` with your actual path.
+
+#### Step 2: Start Backend API (for JMX generation)
+
+```bash
+cd /path/to/Performance-Agent
+python app.py
+```
+
+#### Step 3: Restart VS Code
+
+Press `Cmd+Shift+P` → "Developer: Reload Window"
+
+#### Verify Installation
+
+- Open GitHub Copilot Chat
+- Type: "Analyze the complexity of my last commit"
+
+---
+
+### 🚀 PRODUCTION Deployment
+
+For distributing to end users:
+
+#### Option 1: npm Package (Recommended)
+
+**Publish to npm:**
+
+```bash
+npm publish --access public
+```
+
+**User Installation:**
+
+```bash
+npm install -g @performance-agent/mcp-server
+```
+
+**User Configuration (`.vscode/mcp.json`):**
+
+```json
+{
+  "servers": {
+    "performance-agent": {
+      "command": "performance-agent-mcp",
+      "env": {
+        "PERFORMANCE_AGENT_API_URL": "https://your-backend.azure.com",
+        "PERFORMANCE_AGENT_API_KEY": "your-api-key"
+      }
+    }
+  }
+}
+```
+
+#### Option 2: Standalone Executable
+
+Package as standalone binary (no Node.js required):
+
+```bash
+npm install -g pkg
+pkg . --targets node18-macos-x64,node18-linux-x64,node18-win-x64
+```
+
+Distribute the generated binaries via GitHub Releases.
+
+#### Option 3: Private npm (Enterprise)
+
+```bash
+npm publish --access restricted
+```
+
+Or use Azure Artifacts for private distribution.
+
+---
+
+## 🛠️ Usage
+
+### Tool 1: Generate JMX from Swagger
+
+**In GitHub Copilot Chat:**
+
+```
+Generate JMX script from my swagger.json file
+```
+
+**Parameters:**
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| `repo_name` | string | No | Repository name for the script |
+| `swagger_content` | object | Yes | Swagger/OpenAPI JSON specification |
+
+**Example Response:**
+
+```json
+{
+  "success": true,
+  "jmx_script": "<?xml version=\"1.0\"...",
+  "local_file": "api-test_generated.jmx",
+  "message": "JMX script generated successfully"
+}
+```
+
+---
+
+### Tool 2: Analyze Code Complexity
+
+**In GitHub Copilot Chat:**
+
+```
+Analyze the complexity of my last commit
+```
+
+```
+Check complexity of staged changes with threshold 15
+```
+
+```
+What's the complexity of my uncommitted Python code?
+```
+
+**Parameters:**
+
+| Parameter | Type | Required | Default | Description |
+|-----------|------|----------|---------|-------------|
+| `repo_path` | string | Yes | - | Absolute path to git repo |
+| `mode` | string | No | `last_commit` | `staged`, `last_commit`, `uncommitted`, `commit_sha` |
+| `commit_sha` | string | If mode=commit_sha | - | Specific commit SHA |
+| `threshold` | number | No | `10` | Complexity threshold |
+| `language` | string | No | `all` | `python`, `javascript`, `typescript`, `all` |
+
+**Example Response:**
+
+```json
+{
+  "success": true,
+  "repo_path": "/Users/you/project",
+  "commit_sha": "abc123",
+  "mode": "last_commit",
+  "files_analyzed": 3,
+  "total_functions": 12,
+  "avg_complexity": 4.2,
+  "max_complexity": 15,
+  "threshold_violations": 2,
+  "summary": {
+    "low_risk": 8,
+    "medium_risk": 2,
+    "high_risk": 2,
+    "very_high_risk": 0
+  },
+  "recommendations": [
+    "⚠️  Found 2 function(s) exceeding complexity threshold of 10",
+    "🔴 1. process_request (app.py:145) - Complexity: 15"
+  ]
+}
+```
+
+---
+
+---
+
+### Tool 3: Profile Function Performance
+
+**In GitHub Copilot Chat:**
+
+```
+Profile the performance of my functions
+```
+
+```
+How fast is my process_data function?
+```
+
+```
+Profile functions in my last commit
+```
+
+**Parameters:**
+
+| Parameter | Type | Required | Default | Description |
+|-----------|------|----------|---------|-------------|
+| `repo_path` | string | Yes | - | Absolute path to git repo |
+| `mode` | string | No | `last_commit` | `staged`, `last_commit`, `uncommitted`, `file` |
+| `file_path` | string | If mode=file | - | Specific file to profile |
+| `function_name` | string | No | - | Profile only a specific function |
+| `iterations` | number | No | `10` | Number of times to run each function |
+| `threshold_ms` | number | No | `100` | Time threshold in milliseconds |
+| `language` | string | No | `all` | `python`, `javascript`, `typescript`, `all` |
+
+**Example Response:**
+
+```json
+{
+  "success": true,
+  "repo_path": "/Users/you/project",
+  "mode": "last_commit",
+  "files_profiled": 3,
+  "functions_profiled": 8,
+  "slow_functions": 2,
+  "profiles": [
+    {
+      "file": "utils.py",
+      "function": "process_data",
+      "line": 45,
+      "avg_time_ms": 250.5,
+      "min_time_ms": 180.2,
+      "max_time_ms": 320.1,
+      "performance_level": "slow"
+    }
+  ],
+  "recommendations": [
+    "⚠️ Found 2 slow function(s) exceeding 100ms threshold"
+  ]
+}
+```
+
+---
+
+## 📊 Complexity Thresholds
+
+| Score | Risk Level | Recommendation |
+|-------|------------|----------------|
+| 1-4 | ✅ Low | Good, maintainable code |
+| 5-9 | 👍 Medium | Acceptable |
+| 10-19 | ⚠️ High | Consider refactoring |
+| 20+ | 🔴 Very High | Refactor required |
+
+---
+
+## ⏱️ Performance Thresholds
+
+| Avg Time | Performance Level | Recommendation |
+|----------|-------------------|----------------|
+| < 50ms | 🟢 Fast | Excellent performance |
+| 50-100ms | 🔵 Acceptable | Good performance |
+| 100-200ms | 🟡 Slow | Consider optimization |
+| > 200ms | 🔴 Very Slow | Optimization required |
+
+---
+
+## 💡 Common Use Cases
+
+### Pre-Commit Quality Check
+
+```
+Check complexity of my staged changes
+```
+
+If violations found, refactor before committing.
+
+### Post-Commit Review
+
+```
+Analyze the complexity of my last commit
+```
+
+Create follow-up tasks for high-complexity functions.
+
+### Code Review Helper
+
+```
+Analyze complexity for commit abc123def
+```
+
+Identify functions needing extra attention.
+
+### API Load Testing
+
+```
+Generate JMX script from swagger.json
+```
+
+Run the generated script with JMeter.
+
+### Performance Profiling
+
+```
+Profile the performance of my functions
+```
+
+Identify slow functions and get optimization recommendations.
+
+---
+
+## 🏗️ Architecture
+
+```
+┌─────────────────┐
+│   VS Code       │
+│   GitHub Copilot│
+└────────┬────────┘
+         │ MCP Protocol
+         ▼
+┌─────────────────┐
+│   MCP Server    │  ← This package
+│   (Node.js)     │
+│                 │
+│  ┌───────────┐  │
+│  │ Complexity│  │  ← Runs locally (escomplex/radon)
+│  │ Analyzer  │  │
+│  └───────────┘  │
+│  ┌───────────┐  │
+│  │ Function  │  │  ← Runs locally (timeit/performance.now)
+│  │ Profiler  │  │
+│  └───────────┘  │
+│  ┌───────────┐  │
+│  │ JMX Gen   │──┼──► Backend API (Flask)
+│  └───────────┘  │
+└─────────────────┘
+```
+
+---
+
+## 📁 Project Structure
+
+```
+mcp-server/
+├── src/
+│   ├── index.ts                    # Main MCP server
+│   └── tools/
+│       ├── code-analyzer/
+│       │   ├── index.ts            # Analyzer entry point
+│       │   ├── types.ts            # TypeScript interfaces
+│       │   ├── git-helper.ts       # Git operations
+│       │   └── analyzers/
+│       │       ├── python-analyzer.ts     # Python (radon)
+│       │       └── javascript-analyzer.ts # JS/TS (escomplex)
+│       └── function-profiler/
+│           ├── index.ts            # Profiler entry point
+│           ├── types.ts            # TypeScript interfaces
+│           ├── python-profiler.ts  # Python (timeit)
+│           └── javascript-profiler.ts # JS/TS profiler
+├── dist/                           # Compiled JavaScript
+├── package.json
+├── tsconfig.json
+└── README.md
+```
+
+---
+
+## 🔧 Development
+
+```bash
+# Development mode with hot reload
+npm run dev
+
+# Build TypeScript
+npm run build
+
+# Run tests
+npm test
+```
+
+---
+
+## 🐛 Troubleshooting
+
+### MCP Server Not Loading
+
+1. Check `.vscode/mcp.json` exists with correct path
+2. Restart VS Code: `Cmd+Shift+P` → "Developer: Reload Window"
+3. Check Output panel: View → Output → MCP: performance-agent
+
+### "radon not installed" Warning
+
+```bash
+pip install radon
+```
+
+> The analyzer has a regex fallback, but radon provides better accuracy.
+
+### JMX Generation Fails
+
+Ensure backend API is running:
+
+```bash
+# Local
+python app.py
+curl http://localhost:8000/health
+
+# Production: Check API URL and key in env config
+```
+
+### No Functions Detected
+
+- Ensure files are tracked by git
+- Check file extensions (.py, .js, .ts, .tsx)
+- Try with `mode: "uncommitted"` for untracked files
+
+---
+
+## 🔐 Security
+
+- All complexity analysis runs **locally** - no code sent externally
+- Backend API keys stored in environment variables
+- HTTPS in production for JMX generation API calls
+- No sensitive data logged
+
+---
+
+## 📝 Environment Variables
+
+| Variable | Local | Production | Description |
+|----------|-------|------------|-------------|
+| `PERFORMANCE_AGENT_API_URL` | `http://localhost:8000` | `https://your-api.azure.com` | Backend API URL |
+| `PERFORMANCE_AGENT_API_KEY` | _(empty)_ | Your API key | Authentication key |
+
+---
+
+## 📞 Support
+
+- **Issues**: Create GitHub issue
+- **Logs**: View → Output → MCP: performance-agent
+
+---
+
+## 📄 License
+
+MIT License
+
+---
+
+## ✅ Quick Checklist
+
+### Local Setup
+- [ ] `npm install && npm run build` in mcp-server
+- [ ] Create `.vscode/mcp.json` with local config
+- [ ] Optional: `pip install radon` for Python analysis
+- [ ] Start backend: `python app.py`
+- [ ] Restart VS Code
+- [ ] Test: "Analyze my last commit"
+
+### Production Setup
+- [ ] `npm publish` to npm registry
+- [ ] Deploy backend API to Azure
+- [ ] User installs: `npm install -g @performance-agent/mcp-server`
+- [ ] User configures `.vscode/mcp.json` with production URL/key

package/dist/index.d.ts

@@ -0,0 +1,13 @@
+#!/usr/bin/env node
+/**
+ * Performance Agent MCP Server
+ *
+ * This MCP server provides 3 core tools:
+ * 1. Generate JMX scripts from Swagger/OpenAPI specifications
+ * 2. Analyze code complexity in local git repositories
+ * 3. Profile function execution time and identify slow functions
+ *
+ * Focused on local development and code quality.
+ */
+export {};
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":";AAEA;;;;;;;;;GASG"}