skills.rest
markpitt
markpittCOMMUNITY
Explore all skills by this author →
GitHub

agent-patterns

Orchestrate modular AI agent patterns with deterministic and dynamic workflows.

22|2|Updated Nov 22, 2025
Search Tags:#routing#LLM#patterns#orchestration#agent-patterns#prompt-chaining#dynamic-orchestration
One-click install
npx skills add https://github.com/markpitt/claude-skills --skill agent-patterns-markpitt
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-patterns
Source: https://github.com/markpitt/claude-skills/tree/main/skills/agent-patterns
Command: npx skills add https://github.com/markpitt/claude-skills --skill agent-patterns-markpitt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a modular orchestration framework for AI agent patterns, enabling structured usage of core deterministic patterns and dynamic orchestration patterns to solve complex tasks.

Core Features & Use Cases

  • Pattern Catalog & Quick Reference: Core patterns (Prompt Chaining, Routing, Parallelization) and dynamic patterns (Orchestrator-Workers, Autonomous Agents) organized for fast access.
  • Orchestration Protocol: Phase-based guide to task identification, pattern selection, and cross-language implementation.
  • Language-Agnostic Templates: Provides language-specific templates in resources/language-implementation.md and sample skeletons.
  • Resource Modularity: Separate resources (resources/core-patterns.md, resources/dynamic-orchestration.md, resources/iterative-refinement.md, resources/pattern-combinations.md, resources/language-implementation.md) loaded as needed.

Quick Start

Start by consulting the Quick Reference in SKILL.md to determine the appropriate pattern, load the corresponding resource, and then implement using the language templates in resources/language-implementation.md.

Frequently Asked Questions about agent-patterns

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

FAQPage Schema
How do I orchestrate multiple AI agents to work together on complex tasks?▼

Orchestrate AI agents using core patterns like prompt chaining, routing, and parallelization, or dynamic patterns like orchestrator-workers and autonomous agents. This Skill provides a phase-based protocol to identify your task, select the appropriate pattern, and implement it across Python, TypeScript, C#, Rust, Go, Dart, and C using language-specific templates.

What's the difference between prompt chaining, routing, and parallelization in agent workflows?▼

Prompt chaining sequences outputs as inputs through multiple steps. Routing directs tasks to specialized agents based on conditions. Parallelization runs multiple agents simultaneously. This Skill catalogs all three as core deterministic patterns, with dynamic orchestration patterns for more complex agent interactions.

How do I implement orchestrator-worker or autonomous agent patterns?▼

Use this Skill's orchestration protocol to define task phases, select dynamic orchestration patterns, and implement them with the provided language templates. Resources cover orchestrator-workers for delegated execution and autonomous agents for self-directed workflows, with memory integration and MCP usage guidance.

Can I use these agent patterns across different programming languages?▼

Yes. This Skill provides language-agnostic templates and language-specific implementations for Python, TypeScript, C#, Rust, Go, Dart, and C, enabling you to apply the same orchestration patterns regardless of your tech stack.

How do I choose between deterministic and dynamic agent orchestration patterns?▼

Use deterministic patterns (prompt chaining, routing, parallelization) when task flow is predictable. Use dynamic patterns (orchestrator-workers, autonomous agents, evaluator-optimizer) when agents must adapt or make decisions. This Skill's quick reference guides pattern selection based on your problem structure.

Does this cover iterative refinement in agent workflows?▼

Yes. This Skill includes iterative refinement patterns like evaluator-optimizer for continuous task improvement, covering functional requirements, memory protocols, and cross-language implementations alongside core and dynamic orchestration patterns.

Related Skills

Explore More Software Engineering Skills→
modu-aimodu-ai
1.2k

moai-domain-cli-tool

Coordinate Claude Code workflows by delegating tasks to specialized agents.

Official
Intermediate
PaulKinlanPaulKinlan
2

report-results

Update task statuses with summaries and email output locations to downstream agents.

Community
Basic
EanLeeEanLee

ai-agents-architect

Design autonomous AI agents with tool use, memory, planning, and multi-agent orchestration.

Community
Advanced
dascarddascard
2

workflow-runner

Execute multi-agent YAML workflows with role-based task orchestration.

Community
Advanced