🌊 Alvamind Workflow

A lightweight and flexible workflow automation library for JavaScript/TypeScript projects.

🎯 Introduction

Alvamind Workflow streamlines your development process by providing an elegant way to define and execute multi-step workflows. Whether you prefer YAML configurations or programmatic approaches, Alvamind makes automation simple and maintainable.

✨ Key Features

🔄 Flexible Workflow Definition

• YAML-based configuration for simple, readable workflows
• Programmatic API for dynamic workflow creation
• Mix and match approaches based on your needs

⚡ Powerful Execution Engine

• Real-time execution progress with timing
• Intelligent error handling and recovery
• Skippable steps for optional tasks
• Graceful interruption handling

🔀 Conditional Execution & Dependencies

• Execute steps based on dynamic conditions
• Type-safe dependencies between steps
• Access results from previous steps
• Skip steps when conditions aren't met

🛠️ Developer Experience

• TypeScript-first design
• Minimal configuration required
• Clear, colorful console output
• Comprehensive error messages

🧪 Testing & Development

• Test mode for dry runs
• Promise-based API
• Easy to integrate with CI/CD

🚀 Quick Start

Installation

npm install alvamind-workflow

Simple Example (YAML)

1. Create workflow.yml:

version: "1.0"
name: "Deploy Application"
commands:
  - name: "Build Project"
    command: "npm run build"
  - name: "Run Tests"
    command: "npm test"
    skippable: true
  - name: "Deploy"
    command: "npm run deploy"

2. Run it:

workflow-run

Programmatic Usage

import { createWorkflow } from 'alvamind-workflow';

const workflow = createWorkflow({ name: 'CI Pipeline' })
  .execute('npm run lint', 'Lint Code')
  .execute('npm run test', 'Run Tests', true)
  .execute('npm run build', 'Build Project');

await workflow.run();

📊 Example Output

🐳 Executing workflow: Deploy Application
=====================================
Step 1/3: Build Project
> npm run build
 -> running... 2.5s
[1/3] ✓ Build Project 2.5s

Step 2/3: Run Tests
> npm test
 -> running... 1.2s
[2/3] ✓ Run Tests 1.2s

Step 3/3: Deploy
> npm run deploy
 -> running... 3.1s
[3/3] ✓ Deploy 3.1s
=====================================
✓ Workflow completed in 6.8s

🎛️ Advanced Features

Conditional Execution

Define steps that only run when certain conditions are met:

workflow
  .executeWithId("check", "node -v", "Check Node Version")
  .execute("npm run build", "Build")
  .when(ctx => ctx.getStdout("check").startsWith("v18"), "Node 18 Required")
  .execute("npm run deploy", "Deploy")
  .when(ctx => ctx.getExitCode("build") === 0, "Only deploy if build succeeds")

YAML configuration:

commands:
  - name: "Check Node Version" 
    command: "node -v"
    id: "check"

  - name: "Build"
    command: "npm run build"
    condition: |
      (context) => context.getStdout("check").startsWith("v18")

  - name: "Deploy"
    command: "npm run deploy"
    condition: |
      (context) => context.getExitCode("build") === 0

Dependencies Between Steps

Explicitly define step dependencies:

workflow
  .executeWithId("deps", "npm install", "Install Dependencies")
  .executeWithId("build", "npm run build", "Build")
  .dependsOn("deps")
  .executeWithId("test", "npm test", "Test")
  .dependsOn("build")

YAML configuration:

commands:
  - name: "Install Dependencies"
    command: "npm install"
    id: "deps"

  - name: "Build"
    command: "npm run build"
    id: "build"
    dependsOn: ["deps"]

  - name: "Test"
    command: "npm test"
    dependsOn: ["build"]

Type-Safe Results

Access previous command results with TypeScript type safety:

workflow
  .executeWithId<"check-version">("check-version", "node -v", "Get Version")
  .when(ctx => {
    // TypeScript knows this is valid and provides completion
    const version = ctx.getStdout("check-version");
    return version?.startsWith("v18");
  }, "Verify Version")

Error Handling

try {
  await workflow
    .execute('risky-command', 'Risky Step', true) // skippable
    .execute('must-run', 'Critical Step')         // will fail if error
    .run();
} catch (error) {
  console.error('Workflow failed:', error);
}

Test Mode

await workflow.run({ testMode: true }); // Dry run without executing commands

Interactive Mode

When a command fails, the interactive mode provides options to:

1. Retry the original command
2. Enter a new command to execute
3. Skip the current step (if skippable)
4. Abort the entire workflow

🔧 API Reference

loadWorkflow(path?: string)

Loads a workflow configuration from a YAML file.

• path: (optional) Path to the YAML file. Defaults to workflow.yml.
• Returns: Promise<WorkflowConfig>

runWorkflow(config: WorkflowConfig, options?: RunnerOptions)

Executes a workflow based on the provided configuration.

• config: Workflow configuration object.
• options: (optional) Runner options.
  • testMode: (boolean) Enable test mode to prevent actual command execution.
• Returns: Promise<boolean>

createWorkflow(options?: WorkflowOptions)

Creates a new programmatic workflow builder.

• options: (optional) Workflow options.
  • name: (string) Workflow name.
• Returns: WorkflowBuilder

WorkflowBuilder

A fluent interface for constructing workflows programmatically.

• name(name: string): Sets the workflow name.
• execute<T>(command: string, name: string, skippable?: boolean): Adds a command to the workflow.
• executeWithId<T>(id: T, command: string, name: string, skippable?: boolean): Adds a command with an ID for referencing.
• executeWith(command: string, name: string, callback: (result: { exitCode: number, stdout: string, stderr: string }) => string | undefined, skippable?: boolean): Adds a command with a callback.
• when(condition: (context: WorkflowContext) => boolean | Promise<boolean>, name: string): Adds a condition to the last command.
• dependsOn(...ids: string[]): Defines dependencies for the last command.
• build(): Builds the workflow configuration object (WorkflowConfig).
• run(options?: WorkflowOptions): Runs the workflow.

WorkflowContext

A context object that provides access to previous command results.

• results: Record of all command results.
• getResult(id: string): Gets the full result for a specific command.
• getStdout(id: string): Gets the stdout from a specific command.
• getExitCode(id: string): Gets the exit code from a specific command.
• getStderr(id: string): Gets the stderr from a specific command.

🤝 Contributing

We welcome contributions! Here's how you can help:

• 🐛 Report bugs by opening issues
• 💡 Propose new features
• 📝 Improve documentation
• 🔀 Submit pull requests

📦 What's Inside

alvamind-workflow/
├── src/
│   ├── cli.ts         # CLI implementation
│   ├── index.ts       # Main entry point
│   ├── runner.ts      # Execution engine
│   ├── types.ts       # TypeScript types
│   └── workflow.ts    # Workflow builder
├── test/
│   ├── api.test.ts    # API tests
│   └── conditionals.test.ts # Conditional tests
└── README.md

📄 License

MIT © Alvamind 2025