What planning strategies does GOAP support?

It supports several patterns including Parallel for independent tasks, Sequential for dependent steps, Swarm for similar tasks, and Iterative for progressive refinement.

When should I use the GOAP Agent skill?

Use it for tasks requiring 5+ distinct steps, cross-domain problems spanning multiple modules, or projects that benefit from parallel execution and quality gates.

What is the difference between a Skill and a Task Agent in GOAP?

Skills are instruction sets that provide specialized knowledge and workflows to guide Claude, while Task Agents are autonomous sub-processes that use tools to execute work independently.

How does the GOAP methodology work?

It follows a 6-phase cycle: Analyze goals, Decompose into atomic tasks, Strategize the execution pattern, Coordinate agents, Execute with monitoring, and Synthesize results.

GOAP Agent Orchestrator

Name: GOAP Agent Orchestrator
Author: d-o-hub

byd-o-hub

•

生産性とワークフロー

Plans and executes complex multi-step tasks by coordinating specialized agents through systematic goal decomposition and strategic action planning.

概要

The GOAP (Goal-Oriented Action Planning) Agent skill transforms Claude Code into a sophisticated orchestrator capable of handling intricate, cross-domain technical challenges. It utilizes a structured methodology to analyze high-level goals, map dependencies, and select optimal execution strategies—such as parallel, sequential, or swarm workflows. By intelligently distinguishing between instruction-based Skills for domain expertise and autonomous Task Agents for execution, it ensures that architectural standards are maintained while maximizing efficiency across large-scale refactors, migrations, and feature implementations.

主な機能

Multi-agent coordination across Skills and Task Agents
Advanced execution strategies including Parallel, Swarm, and Hybrid modes
3 GitHub stars
Episodic memory integration for context-aware planning and learning
Dynamic task decomposition with dependency mapping
Integrated quality gates and architectural validation checkpoints

ユースケース

Orchestrating large-scale codebase refactoring requiring multiple validation steps
Implementing complex new features that span across storage, API, and frontend domains
Automating multi-perspective analysis and debugging loops for high-reliability systems

概要

主な機能

Multi-agent coordination across Skills and Task Agents
Advanced execution strategies including Parallel, Swarm, and Hybrid modes
3 GitHub stars
Episodic memory integration for context-aware planning and learning
Dynamic task decomposition with dependency mapping
Integrated quality gates and architectural validation checkpoints

ユースケース

Orchestrating large-scale codebase refactoring requiring multiple validation steps
Implementing complex new features that span across storage, API, and frontend domains
Automating multi-perspective analysis and debugging loops for high-reliability systems