JSPM

snow-flow

1.1.48
  • ESM via JSPM
  • ES Module Entrypoint
  • Export Map
  • Keywords
  • License
  • Repository URL
  • TypeScript Types
  • README
  • Created
  • Published
  • Downloads 4367
  • Score
    100M100P100Q128729F
  • License MIT

Multi-Agent ServiceNow Development Platform - Create, manage, and deploy ServiceNow artifacts using natural language and AI agents

Package Exports

  • snow-flow
  • snow-flow/dist/index.js

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (snow-flow) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

Snow-Flow: Multi-Agent ServiceNow Development Platform ๐Ÿš€

Snow-Flow is a powerful multi-agent AI platform that revolutionizes ServiceNow development through intelligent automation, natural language processing, and autonomous deployment capabilities. Built with 11 specialized MCP (Model Context Protocol) servers, Snow-Flow enables developers to create, manage, and deploy ServiceNow artifacts using simple natural language commands.

๐Ÿ†• What's New in v1.1.45

๐Ÿ” Enhanced Catalog Discovery

  • Fuzzy Search: Find "iPhone 6S" even when searching "iPhone"
  • Smart Variations: Automatically searches related terms (laptop โ†’ notebook, MacBook)
  • Category Filtering: Narrow down results by catalog category

๐Ÿงช Flow Testing Revolution

  • Mock Data Creation: Automatically creates test users and catalog items
  • Approval Simulation: Auto-approves during testing for complete flow validation
  • Execution Monitoring: Real-time tracking with 60-second timeout protection
  • Cleanup Support: Optional removal of all test data after execution

๐Ÿ”— Direct Catalog-Flow Integration

  • Three Linking Methods: Modern flow_catalog_process, legacy workflow, or process_engine
  • Variable Mapping: Map catalog variables to flow inputs with transformations
  • Trigger Control: Configure exactly when flows should execute
  • Test Integration: Create test requests to verify the link works

โšก Performance Improvements

  • Bulk Deployment: Deploy multiple artifacts in a single transaction
  • Parallel Processing: Up to 5x faster with concurrent operations
  • Smart Rollback: Automatic rollback on any deployment failure

๐ŸŒŸ Key Features

๐Ÿค– 11 Specialized MCP Servers

Each server provides autonomous capabilities for different aspects of ServiceNow development:

  1. Deployment MCP - Autonomous widget, flow, and application deployment
  2. Flow Composer MCP - Natural language flow creation with intelligent analysis
  3. Update Set MCP - Professional change tracking and deployment management
  4. Intelligent MCP - AI-powered artifact discovery and editing
  5. Graph Memory MCP - Relationship tracking and impact analysis
  6. Platform Development MCP - Development workflow automation
  7. Integration MCP - Third-party system integration
  8. Operations MCP - Operations and monitoring management
  9. Automation MCP - Workflow and process automation
  10. Security & Compliance MCP - Security auditing and compliance
  11. Reporting & Analytics MCP - Data analysis and reporting

๐ŸŽฏ Core Capabilities

  • Natural Language Processing: Create complex ServiceNow artifacts using plain English/Dutch commands
  • Intelligent Decision Making: Automatically determines optimal architecture (flow vs subflow)
  • Zero Configuration: All values dynamically discovered from your ServiceNow instance
  • Autonomous Deployment: Direct deployment to ServiceNow with automatic error handling
  • Update Set Management: Professional change tracking like ServiceNow pros use
  • Global Scope Strategy: Intelligent scope selection with fallback mechanisms
  • Multi-Agent Coordination: Parallel execution for complex tasks

๐Ÿš€ Quick Start

Prerequisites

  • Node.js 18+ and npm
  • ServiceNow instance with admin access
  • OAuth application configured in ServiceNow

Installation

# Install Snow-Flow globally
npm install -g snow-flow

# Initialize Snow-Flow in your project directory
snow-flow init --sparc

Alternative: Install from source

# Clone the repository
git clone https://github.com/groeimetai/snow-flow.git
cd snow-flow

# Install dependencies
npm install

# Build the project
npm run build

# Link globally (optional)
npm link

Configuration

  1. Create a .env file in the project root:
SNOW_INSTANCE=your-instance.service-now.com
SNOW_CLIENT_ID=your-oauth-client-id
SNOW_CLIENT_SECRET=your-oauth-client-secret
SNOW_USERNAME=your-username
SNOW_PASSWORD=your-password
  1. Set up OAuth in ServiceNow (see SERVICENOW-OAUTH-SETUP.md)

  2. Authenticate with ServiceNow:

snow-flow auth login

๐ŸŽฏ MCP Server Activation (v1.1.25+)

Snow-Flow now includes automatic MCP server activation for Claude Code! During initialization, you'll be prompted to automatically start Claude Code with all 11 MCP servers pre-loaded:

snow-flow init --sparc

# You'll see:
# ๐Ÿš€ Would you like to start Claude Code with MCP servers automatically? (Y/n)
# Press Y to launch Claude Code with all MCP servers ready to use!

The MCP servers are automatically:

  • โœ… Configured with correct paths for global npm installations
  • โœ… Registered in Claude Code's settings
  • โœ… Activated without manual approval steps
  • โœ… Ready to use immediately after initialization

If you need to manually activate MCP servers later:

# For Mac/Linux:
claude --mcp-config .mcp.json .

# For Windows:
claude.exe --mcp-config .mcp.json .

๐Ÿ’ก Usage Examples

Create a Complex Flow with Natural Language

snow-flow sparc "Create an approval workflow for iPhone 6 orders that notifies managers, creates tasks, and updates inventory"

Deploy a Widget Directly to ServiceNow

snow-flow sparc "Create and deploy a widget that shows all critical incidents with real-time updates"

Start a Multi-Agent Swarm for Complex Projects

# Most intelligent features are enabled by default!
snow-flow swarm "Build a complete incident management system with dashboard, workflows, and notifications"

# Default settings:
# โœ… --smart-discovery (true) - Reuses existing artifacts
# โœ… --live-testing (true) - Tests in real-time
# โœ… --auto-deploy (true) - Deploys automatically (safe with update sets)
# โœ… --auto-rollback (true) - Rollbacks on failures
# โœ… --shared-memory (true) - Agents share context
# โœ… --progress-monitoring (true) - Real-time status

# Add --auto-permissions to enable automatic permission escalation
snow-flow swarm "Create enterprise workflow" --auto-permissions

# Disable specific features with --no- prefix
snow-flow swarm "Test workflow" --no-auto-deploy --no-live-testing

Intelligent Artifact Discovery

snow-flow sparc "Find and modify the approval workflow to add an extra approval step for orders over $1000"

Create Flows in Dutch

snow-flow sparc "Maak een flow voor het automatisch toewijzen van incidenten aan de juiste groep op basis van categorie"

๐Ÿ› ๏ธ Advanced Features

Flow vs Subflow Intelligence

Snow-Flow automatically analyzes your requirements and decides whether to create a main flow or break it into reusable subflows:

  • Complexity analysis
  • Reusability assessment
  • Performance optimization
  • Maintainability considerations

Update Set Management

Professional change tracking just like ServiceNow developers use:

# Create a new update set for your feature
snow-flow sparc "Create update set for new approval features"

# All subsequent changes are automatically tracked
snow-flow sparc "Add approval widget to portal"

Global Scope Strategy

Intelligent deployment scope selection:

  • Automatic permission validation
  • Fallback mechanisms for restricted environments
  • Environment-aware deployment (dev/test/prod)

Template Matching

Recognizes common patterns and applies best practices:

  • Approval workflows
  • Fulfillment processes
  • Notification systems
  • Integration patterns

๐Ÿ”ง New MCP Tools (v1.1.44+)

Catalog Item Search with Fuzzy Matching

Find catalog items even when you don't know the exact name:

// In Claude Code with MCP tools
snow_catalog_item_search({
  query: "iPhone",          // Finds iPhone 6S, iPhone 7, etc.
  fuzzy_match: true,       // Intelligent variations
  category_filter: "mobile devices",
  include_variables: true  // Get catalog variables
});

Flow Testing with Mock Data

Test flows without affecting production data:

snow_test_flow_with_mock({
  flow_id: "equipment_provisioning_flow",
  create_test_user: true,      // Auto-creates test user
  mock_catalog_items: true,    // Creates test items
  mock_catalog_data: [
    {
      name: "Test iPhone 6S",
      price: "699.00"
    }
  ],
  simulate_approvals: true,    // Auto-approves
  cleanup_after_test: true     // Removes test data
});

Direct Catalog-Flow Linking

Link catalog items directly to flows for automated fulfillment:

snow_link_catalog_to_flow({
  catalog_item_id: "iPhone 6S",
  flow_id: "mobile_provisioning_flow",
  link_type: "flow_catalog_process",  // Modern approach
  variable_mapping: [
    {
      catalog_variable: "phone_model",
      flow_input: "device_type"
    },
    {
      catalog_variable: "user_department",
      flow_input: "department"
    }
  ],
  trigger_condition: 'current.stage == "request_approved"',
  execution_options: {
    run_as: "system",
    wait_for_completion: true
  },
  test_link: true  // Creates test request
});

Bulk Deployment

Deploy multiple artifacts in a single transaction:

snow_bulk_deploy({
  artifacts: [
    { type: "widget", data: widgetData },
    { type: "flow", data: flowData },
    { type: "script", data: scriptData }
  ],
  transaction_mode: true,  // All-or-nothing deployment
  parallel: true,         // Deploy simultaneously
  dry_run: false
});

๐Ÿ“ Project Structure

snow-flow/
โ”œโ”€โ”€ src/
โ”‚   โ”œโ”€โ”€ mcp/                    # 11 MCP server implementations
โ”‚   โ”œโ”€โ”€ orchestrator/           # Flow composition and intelligence
โ”‚   โ”œโ”€โ”€ strategies/             # Deployment and scope strategies
โ”‚   โ”œโ”€โ”€ api/                    # ServiceNow API integration
โ”‚   โ”œโ”€โ”€ managers/               # Resource and scope management
โ”‚   โ””โ”€โ”€ utils/                  # Utilities and helpers
โ”œโ”€โ”€ .snow-flow/                 # Snow-Flow configuration
โ”œโ”€โ”€ .claude/                    # Claude configuration
โ”œโ”€โ”€ memory/                     # Persistent agent memory
โ””โ”€โ”€ coordination/               # Multi-agent coordination

๐Ÿ”ง Development Commands

# Run tests
npm test

# Run linting
npm run lint

# Type checking
npm run typecheck

# Development mode
npm run dev

# Build for production
npm run build

๐Ÿ“š Documentation

๐Ÿค Contributing

We welcome contributions! Please see our contributing guidelines (coming soon).

๐Ÿ”’ Security

  • All credentials stored securely in environment variables
  • OAuth 2.0 authentication with ServiceNow
  • No hardcoded values - everything discovered dynamically
  • Secure token management with automatic refresh

๐ŸŽฏ Use Cases

For ServiceNow Developers

  • Rapidly prototype flows and workflows
  • Automate repetitive development tasks
  • Ensure consistency across implementations
  • Reduce development time by 80%

For ServiceNow Architects

  • Validate architectural decisions
  • Ensure best practices are followed
  • Analyze impact of changes
  • Optimize performance and maintainability

For ServiceNow Administrators

  • Quick deployments and updates
  • Professional change tracking
  • Automated testing and validation
  • Simplified migration between instances

๐Ÿšฆ Roadmap

  • Visual flow designer integration
  • Enhanced Neo4j graph visualization
  • Multi-instance synchronization
  • AI-powered code review
  • Automated testing framework
  • Performance optimization recommendations

๐Ÿ†• What's New in v1.1.25

Automatic MCP Server Activation ๐ŸŽฏ

  • Interactive Prompt: During snow-flow init --sparc, you're now prompted to automatically start Claude Code with all MCP servers
  • Zero Manual Steps: No more manual MCP approval in Claude Code - servers load automatically using claude --mcp-config
  • Cross-Platform Support: Works on Mac, Linux, and Windows with platform-specific activation scripts
  • Instant Availability: All 11 ServiceNow MCP servers are immediately available in Claude Code after initialization

Previous Updates

  • v1.1.24: Added snow-flow mcp debug command for troubleshooting MCP configurations
  • v1.1.23: Fixed .npmignore to include essential .claude configuration files
  • v1.1.22: Verified global npm installation correctly registers all MCP servers
  • v1.1.20: Added enabledMcpjsonServers to ensure MCP visibility in Claude Code

๐Ÿ“ License

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

๐Ÿ™ Acknowledgments

Built with the power of Claude AI and the ServiceNow platform. Special thanks to the ServiceNow developer community for inspiration and best practices.


Ready to revolutionize your ServiceNow development? Start with snow-flow init --sparc and experience the future of ServiceNow automation! ๐Ÿš€