devtools-bridge-mcp

AI-powered debugging assistant with real-time DevTools integration

GitHubスター

0

ユーザー評価

未評価

フォーク

0

イシュー

0

閲覧数

0

お気に入り

0

README
🌉 DevTools Bridge MCP

Bridge Chrome DevTools with AI assistants for real-time debugging intelligence


✨ What is DevTools Bridge MCP?

DevTools Bridge MCP revolutionizes web debugging by creating a real-time bridge between Chrome DevTools and AI assistants. Instead of manually copying errors and switching between tools, get instant AI-powered analysis of console errors, network issues, and performance bottlenecks—all within your familiar DevTools environment.

🔥 The Problem We Solve

Traditional debugging wastes 2+ hours per day on context switching:

1. 🐛 Error occurs in browser
2. 📋 Copy error message manually  
3. 🔄 Switch to AI chat interface
4. ⌨️  Paste and explain context
5. 🤔 Wait for response
6. 🔄 Switch back to DevTools
7. 🔧 Try suggested solution
8. 🔁 Repeat if unsuccessful

DevTools Bridge MCP eliminates context switching:

1. 🐛 Error occurs in browser
2. 🤖 AI analyzes automatically with full context
3. 💡 Instant solution appears in DevTools
4. 🔧 Apply fix immediately
🥊 Why Choose DevTools Bridge MCP?
Advantage DevTools Bridge MCP GitHub Copilot Chat Cursor/Codeium
Context-Native ✅ Browser DOM, console, network ❌ VS Code only ❌ IDE dependent
Real-time Debugging ✅ Live error capture ❌ Manual copy-paste ❌ Static analysis
Privacy-First ✅ 3-level filtering + on-premises ⚠️ Cloud dependent ⚠️ Data upload
Multi-AI Support ✅ Claude, GPT, Gemini, Local ❌ GPT only ❌ Proprietary models
No IDE Lock-in ✅ Universal browser tool ❌ VS Code required ❌ IDE specific

🎯 Key Features
🤖 Multi-AI Support
  • Claude (Anthropic): Advanced reasoning for complex debugging
  • OpenAI GPT-4: Comprehensive code analysis and suggestions
  • Google Gemini: Fast responses for quick fixes
  • Local LLM: Privacy-focused, on-premises deployment
Real-time Analysis
  • Console Errors: Automatic error capture and AI analysis
  • Network Issues: API debugging with intelligent suggestions
  • Performance Bottlenecks: AI-powered optimization recommendations
  • DOM Mutations: Track and analyze dynamic content changes
🔒 Privacy-First Design
  • 3-Level Privacy Control: Basic → Strict → Paranoid
  • Data Filtering: Automatic removal of sensitive information
  • Self-Hosted Option: Deploy your own MCP server
  • Zero Data Retention: No persistent storage of debugging data
🏢 Enterprise Ready
  • Audit Logging: Complete debugging session history
  • Cost Tracking: Monitor AI API usage and costs
  • Team Settings: Centralized configuration management
  • SSO Integration: Enterprise authentication support

💰 Pricing & Plans
Choose Your Debugging Intelligence Level
Feature Free Starter Growth Enterprise
Monthly Price $0 $15/user $29/user Custom
Seats 1 user Unlimited Unlimited Unlimited
AI Queries 25/month hosted 200/month hosted 1,000/month + overage Unlimited
BYOK Support ✅ Unlimited ✅ Unlimited ✅ Unlimited ✅ Unlimited
Team Features ✅ Basic ✅ Advanced ✅ Enterprise
Support Community Email Priority Dedicated AM
Security Basic Standard Advanced Enterprise
SLA None None 99.5% 99.9%
🎯 Perfect For
  • Free: Individual developers, open-source contributors, students
  • Starter: Freelancers, small teams, personal projects
  • Growth: Growing teams, scale-ups, commercial projects
  • Enterprise: Large organizations, security-conscious teams
🔄 Usage-Based Growth
  • Smart Upgrade Triggers: 75% quota alerts, contextual upgrade prompts
  • 14-Day Growth Trial: Experience advanced features risk-free
  • Overage Billing: $4 per extra 100 queries (Growth tier)
  • Annual Discounts: Save 20% with annual billing
  • No Lock-in: Cancel anytime, export your data
💡 Value Proposition

"Your debugger now thinks with you—without moving to a cloud IDE."

Transform the traditional debugging workflow:

  • Before: Copy error → Switch tools → Wait for analysis → Switch back
  • After: Error occurs → AI analyzes instantly → Solution appears in DevTools

🚀 Start Free Trial💼 Enterprise Demo🎯 14-Day Growth Trial


🎥 Demo

Coming Soon: Watch DevTools Bridge MCP in action

[Demo video will be embedded here once production is complete]

Quick Preview:

  1. 🔧 Install the extension
  2. 🌐 Open any website with console errors
  3. 🤖 Watch AI analyze issues in real-time
  4. 💡 Get instant, actionable solutions

🚀 Quick Start
1. Install Extension

Option A: Chrome Web Store (Recommended)

1. Visit Chrome Web Store
2. Search "DevTools Bridge MCP"  
3. Click "Add to Chrome"
4. Pin extension to toolbar

Option B: Developer Mode

# Clone repository
git clone https://github.com/daideguchi/devtools-bridge-mcp.git
cd devtools-bridge-mcp

# Install dependencies
npm install

# Build extension
npm run build:extension

# Load in Chrome
1. Open chrome://extensions/
2. Enable "Developer mode"
3. Click "Load unpacked"
4. Select dist/extension/ folder
2. Start MCP Server
# Install server dependencies
npm install

# Start the MCP server
npm run dev:server

# Server will start on http://localhost:3001
3. Configure AI Provider
  1. Open Extension Settings: Click the DevTools Bridge icon in your Chrome toolbar
  2. Choose AI Provider: Select Claude, OpenAI, Gemini, or Local LLM
  3. Add API Key: Enter your API key for the chosen provider
  4. Test Connection: Verify the connection is working
4. Start Debugging!
  1. Open DevTools: Press F12 or Ctrl+Shift+I
  2. Find DevTools Bridge Tab: Look for the 🌉 tab in DevTools
  3. Enable Data Collection: Toggle the collection switch
  4. Trigger Errors: Navigate to any website or reproduce bugs
  5. Watch AI Analysis: See real-time insights appear automatically

🏗️ Architecture

DevTools Bridge MCP uses a 4-layer architecture for maximum flexibility and performance:

┌─────────────────────────────────────────────────────────────┐
│                    🎨 UI Layer                              │
│  DevTools Panel • Popup Interface • Options Page           │
└─────────────────────────────────────────────────────────────┘
                                │
┌─────────────────────────────────────────────────────────────┐
│                  🔄 Communication Layer                     │
│     WebSocket • Chrome Extension API • Message Router      │
└─────────────────────────────────────────────────────────────┘
                                │
┌─────────────────────────────────────────────────────────────┐
│                  🧠 Intelligence Layer                      │
│   AI Router • Multi-Provider Support • Context Manager     │
└─────────────────────────────────────────────────────────────┘
                                │
┌─────────────────────────────────────────────────────────────┐
│                   📊 Data Layer                             │
│    Privacy Filter • Data Processor • Audit Logger          │
└─────────────────────────────────────────────────────────────┘
Key Components
  • 🔧 Content Script: Captures debugging data from web pages
  • 🚀 Background Worker: Manages WebSocket connections and routing
  • 🎛️ DevTools Panel: Main debugging interface with 6 specialized tabs
  • 🌐 MCP Server: Node.js server handling AI communications
  • 🤖 AI Router: Intelligent routing between multiple AI providers

📊 Supported Data Types
Data Type Auto-Capture AI Analysis Privacy Filtering
🔴 Console Errors
⚠️ Console Warnings
🌐 Network Requests
📊 Performance Metrics
🏗️ DOM Mutations
💥 Unhandled Rejections
🎯 Custom Events

⚙️ Configuration
Environment Variables
# Server Configuration
MCP_SERVER_PORT=3001
MCP_SERVER_HOST=localhost

# AI Provider Keys
ANTHROPIC_API_KEY=your_claude_key_here
OPENAI_API_KEY=your_openai_key_here  
GOOGLE_AI_API_KEY=your_gemini_key_here

# Privacy & Security
PRIVACY_LEVEL=basic  # basic, strict, paranoid
DATA_RETENTION_DAYS=0  # 0 = no retention
AUDIT_LOGGING=true
Extension Settings

Access via the DevTools Bridge icon in Chrome toolbar:

  • 🤖 AI Configuration: Choose provider, set API keys, configure models
  • 🔒 Privacy & Data: Set filtering level, configure data collection
  • 💰 Cost Tracking: Monitor API usage and set budget alerts
  • ⚙️ Advanced Settings: Custom endpoints, timeout settings, debug mode

🛠️ Development
Prerequisites
  • Node.js: 18+
  • Chrome: 88+ (for Manifest V3 support)
  • TypeScript: 5.0+
  • React: 18+
Setup Development Environment
# Clone and install
git clone https://github.com/daideguchi/devtools-bridge-mcp.git
cd devtools-bridge-mcp
npm install

# Start development servers
npm run dev  # Starts both extension and server in watch mode

# Or run separately
npm run dev:extension  # Extension build + watch
npm run dev:server     # MCP server + nodemon
Project Structure
devtools-bridge-mcp/
├── 📁 src/
│   ├── 📁 extension/          # Chrome extension code
│   │   ├── 📁 background/     # Service worker & WebSocket management
│   │   ├── 📁 content/        # Web page data collection
│   │   ├── 📁 devtools/       # DevTools integration
│   │   ├── 📁 panel/          # Main DevTools panel UI
│   │   ├── 📁 popup/          # Extension popup interface
│   │   └── 📁 options/        # Settings & configuration
│   ├── 📁 server/             # MCP server implementation
│   │   ├── 📁 ai/             # AI provider integrations
│   │   ├── 📁 core/           # WebSocket & MCP handling
│   │   ├── 📁 privacy/        # Data filtering & security
│   │   └── 📁 enterprise/     # Audit logging & cost tracking
│   └── 📁 web/               # Optional web interface
├── 📁 docs/                  # Documentation
├── 📁 scripts/               # Build & deployment scripts
└── 📄 manifest.json          # Chrome extension manifest
Building for Production
# Build extension for distribution
npm run build:extension

# Build MCP server
npm run build:server  

# Create distributable package
npm run package

# The extension will be packaged as devtools-bridge-mcp.zip

🔐 Privacy & Security

DevTools Bridge MCP is designed with privacy-first principles:

🛡️ Data Protection
  • Local Processing: Most analysis happens locally
  • Selective Transmission: Only necessary data sent to AI providers
  • Automatic Filtering: Removes URLs, tokens, keys, and sensitive patterns
  • No Persistence: Zero local data storage by default
🔒 Privacy Levels

Basic Mode

  • Filters obvious secrets (API keys, passwords)
  • Removes query parameters from URLs
  • Safe for most development work

Strict Mode

  • Aggressive URL sanitization
  • Removes stack traces with file paths
  • Filters all request headers
  • Recommended for sensitive projects

Paranoid Mode

  • Maximum privacy protection
  • Local-only analysis when possible
  • Minimal data transmission
  • Enterprise-grade security
🏢 Enterprise Security
  • Audit Trails: Complete logging of all debugging sessions
  • Access Controls: Role-based permissions and API key management
  • Self-Hosted Deployment: Keep all data within your infrastructure
  • Compliance: GDPR, SOC 2, and enterprise security standards

🤝 Contributing

We welcome contributions from the community! Here's how to get involved:

🐛 Reporting Issues
  1. Check existing issues
  2. Use our issue templates
  3. Provide detailed reproduction steps
  4. Include browser version and extension version
💻 Development Contributions
  1. Fork the repository
  2. Create feature branch
    git checkout -b feature/amazing-feature
    
  3. Make your changes
    • Follow our coding standards
    • Add tests for new functionality
    • Update documentation as needed
  4. Commit using Conventional Commits
    git commit -m "feat: add amazing new feature"
    
  5. Push and create Pull Request
    git push origin feature/amazing-feature
    
📋 Development Guidelines
  • Code Style: TypeScript with ESLint + Prettier
  • Testing: Jest for unit tests, Playwright for E2E
  • Documentation: Update README and docs/ for any changes
  • Backwards Compatibility: Maintain API compatibility when possible

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

Third-Party Licenses
  • Chrome DevTools Protocol: BSD License
  • Model Context Protocol: MIT License
  • React: MIT License
  • TypeScript: Apache License 2.0

🙏 Acknowledgments

Special thanks to:

  • 🏢 Anthropic: For Claude AI and inspiration for developer tools
  • 🏢 OpenAI: For GPT models and API accessibility
  • 🏢 Google: For Gemini AI and Chrome DevTools Protocol
  • 🌟 Community: All beta testers and early adopters
  • 🔧 Dependencies: All the amazing open-source libraries that make this possible

📞 Support & Community
🆘 Getting Help
🌟 Stay Updated
  • ⭐ Star this repo to show support and stay notified
  • 👀 Watch releases for new version notifications
  • 🐦 Follow on Twitter: @DevToolsBridge - Latest updates and tips

🚀 Ready to supercharge your debugging workflow?

Install DevTools Bridge MCP Today

Transform debugging from frustration to flow


Made with ❤️ by developers, for developers

Star on GitHub🐦 Share on Twitter

作者情報
Dai

Hi there👋

Japan-Fukuoka

13

フォロワー

45

リポジトリ

0

Gist

0

貢献数

スレッド