sahad0/lazy-AgentX

AgentLazyX1 - Multi-Agent MCP Server

A production-ready Model Context Protocol (MCP) server that provides intelligent automation through specialized agents: Jira, Android Release, and Google Drive integration with LangGraph agents and n8n-style workflow engine.

🏗️ Premium Architecture

This project implements enterprise-grade patterns following LangGraph and MCP SDK best practices:

src/
├── agents/          # LangGraph agents (core intelligence)
│   ├── jira/        # Jira integration agent
│   │   └── JiraAgent.ts
│   ├── android/     # Android release automation agent
│   │   └── AndroidAgent.ts
│   └── google/      # Google Drive integration agent
│       └── GoogleDriveAgent.ts
├── mcp/             # MCP server implementation
│   └── AgentLazyX1MCPServer.ts # Exposes all agents as MCP tools
├── nodes/           # Workflow node implementations
│   ├── BaseNode.ts  # Base class for all nodes
│   ├── jira/        # Jira-specific nodes
│   │   └── JiraNode.ts
│   ├── android/     # Android-specific nodes
│   │   └── AndroidNode.ts
│   └── google/      # Google Drive-specific nodes
│       └── GoogleDriveNode.ts
├── engine/          # Workflow execution engine
│   └── WorkflowEngine.ts
├── workflows/       # Predefined workflow definitions
│   ├── jira/        # Jira workflows
│   │   └── jira-workflow.ts
│   ├── android/     # Android workflows
│   │   └── android-workflow.ts
│   └── google/      # Google Drive workflows
│       └── google-drive-workflow.ts
└── types/           # Type definitions for workflow system
    └── index.ts

🚀 Key Features

• LangGraph-Powered: Core intelligence powered by LangGraph agents
• MCP SDK Integration: Standardized tool exposure via MCP protocol
• n8n-Style Workflows: Extensible node-based workflow system
• Enterprise Ready: Production-grade architecture patterns
• TypeScript: Full type safety and modern development experience
• Cursor Integration: Seamless integration with Cursor IDE
• Multi-Agent Support: Jira, Android Release, and Google Drive agents
• Large File Handling: Streaming uploads for files 100MB+ with progress tracking
• Service Account Authentication: Secure Google Drive integration with service accounts
• Build Automation: Complete Android release pipeline with intelligent error handling

🎯 Architecture Benefits

1. Separation of Concerns

• LangGraph: Handles AI reasoning and workflow orchestration
• MCP SDK: Manages tool exposure and client communication
• Workflow Engine: Executes business logic through nodes

2. Scalability

• Easy to add new agents for different domains
• Extensible node system for complex workflows
• Standardized MCP protocol for tool integration

3. Production Ready

• Proper error handling and logging
• Type-safe implementations
• Modular architecture for easy testing

🔧 Usage

Environment Setup

# Required environment variables

# Jira Configuration
export JIRA_DOMAIN="https://your-domain.atlassian.net"
export JIRA_EMAIL="your-email@company.com"
export JIRA_API_TOKEN="your-jira-api-token"

# Google Drive Configuration (Service Account)
export GOOGLE_SERVICE_ACCOUNT_PATH="./drive-agent-service.json"
export GOOGLE_DRIVE_FOLDER_ID="your-shared-drive-folder-id"

# Google Chat Configuration (Optional - for default space)
export GCHAT_SPACE_ID="spaces/your-google-chat-space-id"

Google Drive Setup

For Google Drive integration, you need to set up a service account:

1. Create Service Account:

   • Go to Google Cloud Console
   • Create a new project or select existing
   • Enable Google Drive API
   • Create a service account
   • Download the JSON key file as drive-agent-service.json

2. Configure Shared Drive:

   • Create a Google Shared Drive (required for service accounts)
   • Share the folder with your service account email
   • Give Editor permissions to the service account
   • Copy the folder ID from the URL

3. Environment Variables:

   export GOOGLE_SERVICE_ACCOUNT_PATH="./drive-agent-service.json"
   export GOOGLE_DRIVE_FOLDER_ID="0ACgT3D-lOrnbUk9PVA"  # Your shared drive folder ID
   

See GOOGLE_DRIVE_SETUP.md for detailed setup instructions.

Google Chat Setup

For Google Chat integration, you need to set up a service account and configure the space:

1. Service Account Setup (same as Google Drive):

   • Use the same service account JSON file (service.json) as Google Drive
   • Ensure the service account has Google Chat API access

2. Add Bot to Google Chat Space:

   • Open your Google Chat space
   • Click on the space name → "Manage webhooks and apps"
   • Add the service account email as a bot
   • Give it appropriate permissions

3. Get Space ID:

   • Open the Google Chat space in your browser
   • Copy the space ID from the URL: https://chat.google.com/room/XXXXXXXXX
   • The space ID format is: spaces/XXXXXXXXX

4. Environment Variables:

   export GCHAT_SPACE_ID="spaces/your-google-chat-space-id"
   

5. Usage:

   • With default space: Send message 'Hello World'
   • With specific space: Send message 'Hello World' to space spaces/abc123

Running the System

# Development mode with hot reload
yarn dev

# Production MCP server
yarn mcp

# Production build and start
yarn build
yarn start:prod

# LangGraph API server (alternative deployment)
yarn langgraph:start

# Development LangGraph server
yarn langgraph:dev

Production Deployment

1. Environment Setup:

   cp env.example .env
   # Edit .env with your production values
   export NODE_ENV=production
   

2. Build and Deploy:

   yarn build
   yarn start:prod
   

3. Health Monitoring:

   • Health checks run every 5 minutes in production
   • Monitor logs for uptime and error messages
   • All agents include basic error handling and logging

Cursor Integration

The MCP server exposes multiple specialized tools:

🤖 Available Agents & APIs

1. Jira Agent

Tools: jira_agent, jira_advanced_search

Capabilities:

• Intelligent Jira Queries: Natural language processing of Jira requests
• Workflow Execution: n8n-style workflow processing
• AI-Enhanced Responses: Context-aware ticket summaries
• Multi-format Support: Handles ticket keys, URLs, and descriptions
• Advanced Search: Filter by assignee, status, priority, issue type, project

API Parameters:

// jira_agent
{ query: string }

// jira_advanced_search
{
  text?: string,           // Search text
  assignee?: string,       // Filter by assignee
  status?: string,         // Filter by status
  priority?: string,       // Filter by priority
  issueType?: string,      // Filter by issue type
  project?: string,        // Filter by project
  maxResults?: number      // Max results (default: 10)
}

2. Android Release Agent

Tool: android_release_agent

Capabilities:

• Build Automation: Automated Android release builds
• Error Analysis: Intelligent error detection with solutions
• Progress Tracking: Real-time build progress monitoring
• Clean Build Management: Smart cache clearing and build folder management
• Extended Timeout: Optimized for low-end devices

API Parameters:

{
  command: string,                    // Required: Command to execute
  projectPath?: string,               // Optional: Project path
  runGradleClean?: 'yes'|'no'|'auto' // Gradle clean control
}

Special Commands:

• android:release - Full Android release build workflow
• yarn build - Standard build command
• yarn android:release - Android-specific release

3. Google Chat Agent

Tool: google_chat_agent

Capabilities:

• Text Messaging: Send messages to Google Chat spaces
• User Mentions: Tag specific users by name or email
• URL Attachments: Include links with custom display text
• Default Space: Configure a default space for easy messaging
• Enhanced Features: Support for @all mentions and thread replies

API Parameters:

{
  query: string; // Natural language query for sending messages
}

Query Examples:

// Simple message to default space
"Send message 'Hello World'";

// Message to specific space
"Send message 'Hello World' to space spaces/abc123";

// Message with user mentions
"Send message 'Hello team' tag users 'john, jane'";

// Message with URL
"Send message 'Check this out' url 'https://example.com' url text 'Example Link'";

// Message with @all mention
"Send message 'Important update' tag all";

4. Google Drive Agent

Tool: google_drive_upload

Capabilities:

• Large File Uploads: Streaming uploads for files 100MB+ with resumable upload
• Progress Tracking: Real-time upload progress with speed and ETA
• Memory Efficient: Uses streaming to avoid loading entire files into memory
• Service Account Auth: Secure authentication with Google service accounts
• Public Sharing: Automatic public sharing permissions
• File Replacement: Smart handling of existing files

API Parameters:

{
  filePath: string,        // Required: Local file path
  folderId?: string,       // Optional: Google Drive folder ID
  fileName?: string,       // Optional: Custom file name
  enableProgress?: boolean // Optional: Enable progress tracking
}

Features:

• Automatic Optimization: Chooses simple vs resumable upload based on file size
• Progress Tracking: Real-time upload progress with speed and ETA
• Public URLs: Returns both direct and sharing links
• Error Handling: Comprehensive error messages with solutions

🏭 Production Deployment

Option 1: Direct MCP Server

yarn mcp

• Runs as stdio MCP server
• Direct integration with Cursor
• Best for development and single-user scenarios

Option 2: LangGraph API Server

yarn langgraph:start

• Runs as HTTP server with /mcp endpoint
• Supports multiple clients
• Better for production and team environments

🔌 Extending the System

Adding New Agents

1. Create agent class in src/agents/
2. Implement workflow integration
3. Register in MCP server

Adding New Workflow Nodes

1. Extend BaseNode class
2. Implement getDefinition() and execute() methods
3. Register in WorkflowEngine

Adding New Workflows

1. Create workflow definitions in src/workflows/
2. Define node connections and parameters
3. Integrate with agents

🚀 Future Enhancements

• Multi-Agent Orchestration: Coordinate multiple specialized agents
• Advanced Workflows: Complex branching and parallel execution
• Visual Workflow Editor: n8n-style drag-and-drop interface
• Enterprise Integrations: GitHub, Slack, email, CRM systems
• Real-time Monitoring: Workflow execution tracking and analytics
• Custom Node Marketplace: Community-driven node ecosystem
• Google Drive List/Download: File listing and download capabilities
• Android Build Variants: Support for different build flavors and variants
• Jira Webhook Integration: Real-time ticket updates and notifications

🛠️ Development

# Install dependencies
yarn install

# Format code
yarn format

# Build TypeScript
yarn build

# Run tests (when implemented)
yarn test

📚 Architecture Patterns

This implementation follows these premium patterns:

1. Agent-First Design: LangGraph agents as the core intelligence layer
2. Protocol Standardization: MCP SDK for tool exposure
3. Workflow Abstraction: n8n-style node system for business logic
4. Type Safety: Full TypeScript implementation
5. Modular Architecture: Clear separation of concerns
6. Production Readiness: Error handling, logging, and monitoring

📄 License

MIT