workflow-orchestrator

DAG-based workflow automation with parallel task execution and dependency management

View on GitHub
Author Jeremy Longshore
Namespace @jeremylongshore/claude-code-plugins-plus
Category automation
Version 1.0.0
Stars 1,193
Downloads 2
self.md verified
Table of content

DAG-based workflow automation with parallel task execution and dependency management

Installation

npx claude-plugins install @jeremylongshore/claude-code-plugins-plus/workflow-orchestrator

Contents

Folders: servers, skills

Files: LICENSE, README.md, package.json, server.json, tsconfig.json

Documentation

DAG-based workflow orchestration with parallel execution and run history

Automate complex multi-step workflows with dependency management, parallel execution, and comprehensive run tracking.

Features

Installation

/plugin install workflow-orchestrator@claude-code-plugins-plus

4 MCP Tools

1. create_workflow

Define workflow with tasks and dependencies.

{
  "name": "Build and Deploy",
  "tasks": [
    {"id": "lint", "name": "Lint code", "command": "npm run lint", "dependencies": []},
    {"id": "test", "name": "Run tests", "command": "npm test", "dependencies": ["lint"]},
    {"id": "build", "name": "Build", "command": "npm run build", "dependencies": ["test"]},
    {"id": "deploy", "name": "Deploy", "command": "npm run deploy", "dependencies": ["build"]}
  ]
}

2. execute_workflow

Run workflow with parallel execution.

{
  "workflowId": "wf_1234567890",
  "parallel": true
}

3. get_workflow

Get workflow status and task details.

{
  "workflowId": "wf_1234567890"
}

4. list_workflows

List all workflows with optional filtering.

{
  "status": "completed"
}

Quick Start

// 1. Create workflow
const workflow = await create_workflow({
  name: "CI/CD Pipeline",
  tasks: [
    { id: "lint", name: "Lint", command: "npm run lint", dependencies: [] },
    { id: "test", name: "Test", command: "npm test", dependencies: ["lint"] },
    { id: "build", name: "Build", command: "npm run build", dependencies: ["test"] }
  ]
});

// 2. Execute workflow
const result = await execute_workflow({
  workflowId: workflow.workflowId,
  parallel: true
});

// 3. Check status
const status = await get_workflow({
  workflowId: workflow.workflowId
});

DAG Execution

Tasks execute based on dependency graph:

  lint
  test ← test:integration (parallel)
    ↓        ↓
  build ←───┘
  deploy

Independent tasks (test + test:integration) run in parallel.

Use Cases

  1. CI/CD Pipelines - Automated build, test, deploy
  2. Data Pipelines - ETL workflows with dependencies
  3. Deployment Automation - Multi-stage deployments
  4. Testing Workflows - Parallel test execution
  5. Batch Processing - Complex job orchestration

License

MIT License


Made with ️ by Intent Solutions

Source

View on GitHub

Tags: automation mcpworkflowdagautomationorchestrationcicdparallel-execution