skills.rest
fartmonarch
fartmonarchCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

workflow-runner

Execute YAML-defined multi-role workflows with dependency-aware sequencing and parallel steps.

1|Updated Apr 7, 2026
Search Tags:#automation#workflow#multi-agent#orchestration#yaml#dependency planning#role execution
One-click install
npx skills add https://github.com/fartmonarch/RenderMind --skill workflow-runner-fartmonarch
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: workflow-runner
Source: https://github.com/fartmonarch/RenderMind/tree/main/.claude/skills/workflow-runner
Command: npx skills add https://github.com/fartmonarch/RenderMind --skill workflow-runner-fartmonarch

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the friction of setting up external orchestration tools when you need a multi-step, multi-role workflow completed inside the current AI session. It lets you parse a YAML workflow, resolve inputs, execute each role in order, and preserve outputs without leaving the chat.

Core Features & Use Cases

  • YAML workflow parsing: Reads agency-orchestrator style workflow definitions and extracts inputs, steps, dependencies, and execution metadata.
  • Role-based execution: Loads each role definition from the agent directory and performs the task using that role's system instructions and communication style.
  • Dependency-aware planning: Topologically sorts steps, runs independent steps in parallel when possible, and keeps shared outputs available for later steps.
  • Persistent run artifacts: Saves step outputs, summaries, and metadata into a structured output folder for review and reuse.
  • Use cases: Running a product review workflow, coordinating research and writing roles, or executing a custom multi-agent process from a provided YAML file.

Quick Start

Ask the skill to run your YAML workflow file and provide any required input values so it can execute the steps and save the results.

Frequently Asked Questions about workflow-runner

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

FAQPage Schema
How do I run a multi-agent workflow in chat without external orchestration tools?▼

To define a multi-role workflow, you write a YAML file specifying inputs, steps, dependencies, and execution metadata. The runner parses this YAML to load roles, resolve inputs, and execute tasks in dependency-aware order.

Can I execute parallel steps with dependency-aware planning in a YAML workflow?▼

Yes, dependency-aware planning topologically sorts your workflow steps to run independent steps in parallel. Shared outputs from completed steps are kept available for later dependent steps.

Do I need external API keys to run role-based tasks inside an AI chat session?▼

No, you do not need external API keys to run role-based tasks. The workflow runner executes YAML-defined multi-role workflows entirely inside the current AI chat session without external dependencies.

What is the best way to coordinate research and writing roles in a single AI session?▼

The best way to coordinate research and writing roles is using a YAML workflow that defines each role's system instructions and communication style. The runner orchestrates these roles, executing tasks in order and preserving outputs.

How are workflow step outputs and metadata saved for review after execution?▼

Workflow step outputs, summaries, and metadata are saved into a structured output folder as persistent run artifacts. This ensures all execution results remain available for later review and reuse.

What happens if a step fails during multi-role YAML workflow execution?▼

The workflow runner applies failure-aware step handling during execution. If a step fails, the system manages the error according to the workflow's defined logic to preserve the overall execution state.

Related Skills

Explore More Software Engineering Skills→
majiayu000majiayu000
12

sage-prompt-engineering

Render reusable system prompts from templates with variable injection.

Community
Basic
clinecline
19

artifact-management

Organize data analysis outputs into dated folders with metadata and safety guardrails.

Official
Intermediate
xlelibertyxleliberty

agent-designer

Designs scalable multi-agent system architectures and generates tool schemas.

Community
Advanced
yashurathodyashurathod

V3 Swarm Coordination

Coordinate a 15-agent hierarchical swarm for Claude-Flow v3 implementation.

Community
Advanced