setup-workflow

Set up DAG-based workflow configuration and project scaffolding for the flowai-workflow engine.

Updated Mar 7, 2026

Search Tags:#automation #workflow #claude #prompts #deno #dag #hitl

Ready to run

npx skills add https://github.com/korchasa/flowai-workflow --skill setup-workflow-korchasa

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: setup-workflow
Source: https://github.com/korchasa/flowai-workflow/tree/main/.claude/skills/setup-workflow
Command: npx skills add https://github.com/korchasa/flowai-workflow --skill setup-workflow-korchasa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a clear, reproducible way to add and configure the flowai-workflow DAG engine in a project so teams can define, validate, and run multi-agent pipelines without ad hoc folder layouts or inconsistent prompt handling.

Core Features & Use Cases

Project scaffolding: Creates the .flowai-workflow layout, run artifact directories, and common prompt locations to standardize workflow projects.
YAML workflow composition: Guides authors to define nodes, phases, loop and human gates, validation rules, and defaults for consistent DAG execution.
Runtime and validation readiness: Covers Deno or prebuilt binary usage, Claude/OpenCode runtime settings, validation rules, and dry-run/resume patterns for robust execution.

Quick Start

Run flowai-workflow with --config .flowai-workflow/workflow.yaml to validate and execute your workflow.

Frequently Asked Questions about setup-workflow

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I set up a DAG-based multi-agent workflow configuration?▼

To set up a DAG-based workflow, this Skill scaffolds a standardized .flowai-workflow directory and generates a validated workflow.yaml file defining agent nodes, prompt fragments, and human-in-the-loop gates for the flowai-workflow engine.

What is the best way to manage prompt fragments for AI pipelines?▼

Managing prompt fragments for AI pipelines is standardized by creating common prompt file conventions within the workflow scaffolding, ensuring reproducible multi-agent execution without ad hoc folder layouts or inconsistent prompt handling.

Does the flowai-workflow engine support human-in-the-loop and iterative looped pipelines?▼

Yes, human-in-the-loop and iterative looped pipelines are supported. The generated workflow.yaml configuration allows you to define specific node phases, human gates, and validation rules to control DAG execution flow.

Can I run flowai-workflow pipelines with Deno or prebuilt binaries?▼

You can run flowai-workflow pipelines using either Deno or prebuilt binary runtimes. The project scaffolding includes optional helper scripts and runtime settings configured for Claude or OpenCode execution environments.

How do I validate and execute a YAML-defined agent workflow?▼

To validate and execute a YAML-defined agent workflow, run the flowai-workflow engine with the --config .flowai-workflow/workflow.yaml flag, leveraging built-in validation rules, dry-run capabilities, and resume patterns for robust pipeline execution.