skills-workflows

Creates sequenced skill workflows with artifact-based state handoff and preprocessing via command syntax.

6|1|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/outfitter-dev/outfitter --skill skills-workflows
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skills-workflows
Source: https://github.com/outfitter-dev/outfitter/tree/main/plugins/fieldguides/skills/skills-workflows
Command: npx skills add https://github.com/outfitter-dev/outfitter --skill skills-workflows

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a robust framework for designing and implementing complex, multi-step workflows where skills pass data and state between each other using artifacts, ensuring deterministic and auditable execution.

Core Features & Use Cases

  • Skill Chaining: Define sequences of skills that execute in order, with the output of one feeding into the next.
  • Artifact-Based State Handoff: Use standardized file artifacts for passing complex data between skills, avoiding context window limitations and ensuring persistence.
  • Workflow Isolation: Control context inheritance (fork vs. inherit) for cleaner execution and parallel processing.
  • Preprocessing: Use !command syntax to inject live system data directly into prompts before they are sent to the model.
  • Use Case: Building an automated code review pipeline where one skill generates a summary, another performs a security audit, and a third generates a report, all passing their findings via artifacts.

Quick Start

Use the skills-workflows skill to design a multi-skill pipeline for code review, starting with a triage skill and ending with a shipping skill.

Frequently Asked Questions about skills-workflows

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

FAQPage Schema
How do I pass complex data between sequential skills in a pipeline?

Yes, you can inject live system data into prompts before model processing using the `!command` syntax. This preprocessing capability enables deterministic context injection directly into your workflow steps.

What is the best way to orchestrate a multi-step code review pipeline?

The best way to orchestrate a multi-step code review pipeline is by chaining skills sequentially. Define execution order so one skill's output feeds the next, passing findings via artifacts for a fully auditable workflow.

How does workflow isolation manage context inheritance during skill orchestration?

Workflow isolation manages context inheritance through fork and inherit controls. This allows cleaner execution and parallel processing by strictly controlling what context is passed between sequential skill executions.

Do I need artifact storage to build skill pipelines with state handoff?

Yes, artifact storage is required for state handoff in skill pipelines. Standardized file artifacts provide the persistence mechanism needed to pass complex data between skills without hitting context window limits.

Why use artifact-based handoff instead of direct context passing for multi-skill workflows?

Artifact-based handoff is used instead of direct context passing to avoid context window limitations and ensure state persistence. It provides a robust framework for auditable data transfers between sequential skill executions.