sft-coordination-patterns

Select multi-agent coordination patterns for Claude Code skill workflows.

Updated Mar 30, 2026
One-click install
npx skills add https://github.com/steltz/skill-factory --skill sft-coordination-patterns
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sft-coordination-patterns
Source: https://github.com/steltz/skill-factory/tree/main/plugins/skill-factory-toolkit/skills/sft-coordination-patterns
Command: npx skills add https://github.com/steltz/skill-factory --skill sft-coordination-patterns

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Pattern-guided selection and application of multi-agent coordination structures to design robust skill workflows, reducing guesswork and misaligned interactions.

Core Features & Use Cases

  • Reference table and detailed guidance on five coordination patterns (Orchestrator-Subagent, Generator-Verifier, Agent Teams, Message Bus, Shared State)
  • Worked examples and design principles that illustrate how to structure prompts, tasks, and data flows for each pattern
  • Decision guidance for when to apply each pattern and how to document coordination contracts and termination conditions

Quick Start

Install or load the coordination-patterns toolkit and review the pattern details to begin applying them in your skill design.

Frequently Asked Questions about sft-coordination-patterns

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

FAQPage Schema
What are the main multi-agent coordination patterns for structuring workflows?

The main multi-agent coordination patterns are orchestrator-subagent, generator-verifier, agent teams, message bus, and shared state. Each provides a distinct structural template for routing tasks, managing data flows, and handling termination conditions across multiple agents.

How do I choose the right coordination pattern for a multi-agent skill design?

To choose the right coordination pattern, evaluate explicit selection criteria against your workflow requirements. Compare the scale, task dependency, and verification needs of your workflow against the reference guidance to identify whether an orchestrator-subagent or shared state model fits best.

When should I use a generator-verifier pattern instead of an orchestrator-subagent model?

Use a generator-verifier pattern when your workflow requires explicit output validation and quality gating before proceeding. An orchestrator-subagent model is better suited for dynamic task decomposition and parallel execution where strict sequential verification is not the primary constraint.

How do I document coordination contracts and termination conditions for agent teams?

Document coordination contracts and termination conditions by applying the provided template guidance and phase maps. This structures the prompts, tasks, and data flow boundaries that define when and how agent teams safely conclude their interactions.

Can I use these coordination templates for the planning phase of a new skill?

Yes, these coordination templates are explicitly applicable to the design and planning phases of new skills. They provide reference material and decision guidance to structure prompts and coordination artifacts before implementation begins.