workflow-patterns

Standardize multi-agent workflow orchestration with sequential, parallel, fork-join, iteration, and conditional routing patterns.

2|Updated Oct 19, 2025
One-click install
npx skills add https://github.com/masanao-ohba/claude-manifests --skill workflow-patterns-masanao-ohba
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: workflow-patterns
Source: https://github.com/masanao-ohba/claude-manifests/tree/main/skills/generic/workflow-patterns
Command: npx skills add https://github.com/masanao-ohba/claude-manifests --skill workflow-patterns-masanao-ohba

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinating multiple AI agents can be complex and error-prone without standardized patterns, and teams require reusable guidance to ensure consistency.

Core Features & Use Cases

  • Sequential Chain Pattern: agents execute in strict order, with clear dependency chaining.
  • Parallel Execution Pattern: enables simultaneous work for independent tasks and aggregated results.
  • Fork-Join Pattern: handles concurrent tasks with synchronization on join.
  • Iteration Pattern: repeats cycles until success or predefined limits.
  • Conditional Routing Pattern: routes tasks based on evaluation outcomes.
  • Workflow Templates: ready-to-use templates that encode end-to-end processes.

Quick Start

To use this skill, apply the provided patterns to your agent orchestration design, refer to the core examples, and adapt to your project workflow.

Frequently Asked Questions about workflow-patterns

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

FAQPage Schema
What are the standard patterns for multi-agent workflow orchestration?

Standard multi-agent workflow orchestration patterns include sequential chain, parallel execution, fork-join, iteration, and conditional routing, each providing deterministic coordination for complex task execution.

How do I coordinate parallel AI agents to aggregate independent task results?

To coordinate parallel AI agents, apply the parallel execution pattern to enable simultaneous work for independent tasks and aggregate the results, ensuring efficient multi-agent coordination without dependency conflicts.

When should I use fork-join vs sequential chain patterns in agent workflows?

Use the fork-join pattern for concurrent tasks requiring synchronization on join, and apply sequential chain patterns when agents must execute in strict order with clear dependency chaining.

How to route tasks across multiple agents based on conditional evaluation outcomes?

Route tasks by applying the conditional routing pattern, which directs workflow paths based on evaluation outcomes to ensure deterministic and traceable task execution across multiple agents.

Are there ready-to-use templates for multi-agent workflow design and testing?

Yes, ready-to-use workflow templates encode end-to-end processes for design, development, and testing scenarios, providing reusable guidance to ensure consistency and reduce coordination errors.