AI Sprint

AI-powered sprint management and SDLC tracking with automatic IDE integration. Features intelligent session tracking, test reporting, and seamless BizOps Portal synchronization for Claude, Cursor, VS Code, and more.

✨ Features

• 🔍 IDE Auto-Detection: Automatically detects Claude Code, Cursor, Windsurf, VS Code
• ⚡ Zero Configuration: One command setup with intelligent defaults
• 📝 Hybrid Integration: Use both MCP tools and IDE commands (Claude Code)
• 📊 Real-time Tracking: Session files, git commits, test results
• 🔄 Live Synchronization: Automatic Portal updates as you work
• 📈 Sprint Management: Comprehensive sprint planning and tracking

🚀 Quick Start

Automatic Setup (Recommended)

# Install globally
npm install -g ai-sprint

# Run setup in your project directory
ai-sprint setup

The setup will:

1. Detect your IDE(s) automatically
2. Configure MCP integration
3. Generate command files (for Claude Code)
4. Set up project tracking

Manual Installation

npm install ai-sprint

Then add to your IDE configuration:

Claude Code

{
  "mcpServers": {
    "ai-sprint": {
      "command": "npx",
      "args": ["ai-sprint"],
      "env": {
        "AUTO_CONFIGURE": "true",
        "IDE_TYPE": "claude"
      }
    }
  }
}

Cursor

{
  "mcpServers": {
    "ai-sprint": {
      "command": "npx",
      "args": ["ai-sprint"],
      "env": {
        "AUTO_CONFIGURE": "true",
        "IDE_TYPE": "cursor"
      }
    }
  }
}

📱 IDE-Specific Features

Claude Code

Claude Code users get both commands and MCP tools:

Commands (in .claude/commands/)

• /session-start [type] [name] - Start a new development session
• /session-update [checkpoint] [notes] - Update session progress
• /session-end - Complete session with metrics
• /sprint-create [type] [name] [duration] - Create new sprint
• /test-report [type] [passed] [failed] - Submit test results

MCP Tools

• configure - Set up project tracking
• sync_sessions - Sync all sessions to Portal
• sync_git_history - Sync git commits
• submit_test_results - Report test results
• watch - Start file monitoring

Other IDEs

Use MCP tools directly through your IDE's interface.

🛠️ Usage

1. Configure the Tracker

// Configure for your project
await configure({
  projectId: "your-project-uuid",
  projectPath: "/path/to/your/project",
  apiEndpoint: "https://portal.unblockd.com",
  apiKey: "optional-api-key",
  watchPatterns: ["sessions/*.md", "docs/sprint-*.md"]
});

2. Sync Sessions

// Sync all session files
await sync_sessions();

3. Sync Git History

// Sync git commits
await sync_git_history({
  branch: "main",
  since: "2024-01-01"
});

4. Submit Test Results

// Submit test run results
await submit_test_results({
  sessionId: "session-uuid",
  testType: "unit",
  passed: 245,
  failed: 3,
  skipped: 12,
  coverage: 87.5,
  duration: 12340
});

5. Start Watching

// Start watching for file changes
await watch();

Session File Format

The tracker expects session files to follow this naming convention:

sessions/YYYY-MM-DD-HHMM-[type]-[descriptive-name].md

Where type is one of:

• feature - Feature development
• bug - Bug fixes
• maintenance - Maintenance tasks
• research - Research and investigation
• emergency - Emergency fixes

Example Session File

# Feature: Add User Authentication

## Goals
- Implement secure user authentication
- Add role-based access control
- Create login/logout UI

## Acceptance Criteria
- Users can register with email/password
- Users can login and logout
- Sessions persist across page refreshes
- Unauthorized users are redirected to login

## Test Results
- Passed: 45
- Failed: 0
- Skipped: 3
- Coverage: 92.3%

## Performance Metrics
- API Response Time: 120ms
- Page Load Time: 450ms
- Component Render Time: 35ms

API Integration

The MCP server sends data to these BizOps Portal endpoints:

• POST /api/sdlc/sessions - Create/update sessions
• POST /api/sdlc/git-sync - Sync git commits
• POST /api/sdlc/test-runs - Submit test results
• POST /api/sdlc/sprints - Update sprint data

🔧 CLI Commands

Setup and Configuration

# Interactive setup
ai-sprint setup

# Non-interactive setup with options
ai-sprint setup --project-id <uuid> --endpoint <url> --api-key <key> --no-interactive

# Show current configuration
ai-sprint info

# Test API connection
ai-sprint test-connection

Server Management

# Start the MCP server manually
ai-sprint start

# Start with debug logging
ai-sprint start --debug

🏗️ Development

# Clone the repository
git clone https://github.com/unblockd/ai-sprint
cd ai-sprint

# Install dependencies
npm install

# Build the project
npm run build

# Run in development mode
npm run dev

Project Structure

├── bin/                    # CLI executables
│   ├── ai-sprint       # Main CLI
│   └── ai-sprint-setup # Setup wizard
├── src/                   # Source code
│   ├── index.ts          # MCP server
│   ├── ide-detector.ts   # IDE detection logic
│   └── configurator.ts   # Auto-configuration
├── dist/                 # Compiled output
└── package.json         # Package configuration

🔐 Environment Variables

• BIZOPS_API_ENDPOINT - Portal API endpoint
• BIZOPS_API_KEY - API authentication key
• PROJECT_ID - Override project ID
• IDE_TYPE - Force specific IDE type
• AUTO_CONFIGURE - Enable auto-configuration
• DEBUG - Enable debug logging

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

📄 License

MIT - See LICENSE file for details

🔗 Links

• BizOps Portal
• MCP Documentation
• Issue Tracker