@tsflow/flow-engine

Enterprise TypeScript workflow engine with state machines, parallel execution, saga patterns, and YAML DSL.

Installation

npm install @tsflow/flow-engine

Quick Start

import { FlowEngine, StateMachineConfig } from '@tsflow/flow-engine';

const config: StateMachineConfig = {
  id: 'order-flow',
  initialState: 'pending',
  states: {
    pending: {
      name: 'pending',
      transitions: [
        { event: 'APPROVE', to: 'approved' },
        { event: 'REJECT', to: 'rejected' }
      ]
    },
    approved: { name: 'approved', type: 'final' },
    rejected: { name: 'rejected', type: 'final' }
  }
};

const engine = new FlowEngine(config);
const flow = await engine.start({ context: { orderId: '12345' } });

await engine.execute(flow.flowId, { event: 'APPROVE' });
console.log('Order approved!');

Features

• 🎯 State Machine Core - Robust state machine with transitions, guards, and actions
• ⚡ Parallel States - Execute multiple regions concurrently
• 🏗️ Hierarchical States - Nested state machines for complex workflows
• 🎭 Saga Pattern - Built-in compensation/rollback for distributed transactions
• 🔁 Sub-Flows - Compose workflows from reusable components
• 🔌 Middleware - Hooks for logging, metrics, observability
• ♻️ Retry Logic - Automatic retry with exponential backoff
• 🔐 Idempotency - Prevent duplicate executions
• ✅ Validation - Schema-based context validation
• 📝 YAML DSL - Define workflows declaratively
• 🏷️ Type-Safe - Full TypeScript support
• 🧪 Well Tested - 162+ comprehensive tests

Documentation

Complete documentation available at:

• Getting Started
• Advanced Patterns
• YAML Workflows
• Examples
• API Reference

Use Cases

• E-Commerce (order processing, payments, fulfillment)
• User Management (registration, verification, onboarding)
• Content Management (approval workflows, publishing)
• DevOps (CI/CD, infrastructure provisioning)
• Finance (loan applications, KYC, transactions)
• Customer Support (ticket management, escalation)
• IoT (device provisioning, commissioning)

License

MIT

Repository

https://github.com/sjefvanleeuwen/tsFlow