agentic-workflow-guide

Catalog agent workflow patterns and governance rules for orchestration.

24|4|Updated Dec 24, 2025

Search Tags:#workflow #agent #orchestration #design-principles #handoffs #context-engineering #sub-agent

Ready to run

npx skills add https://github.com/aktsmm/Agent-Skills --skill agentic-workflow-guide

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agentic-workflow-guide
Source: https://github.com/aktsmm/Agent-Skills/tree/main/agentic-workflow-guide
Command: npx skills add https://github.com/aktsmm/Agent-Skills --skill agentic-workflow-guide

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps teams design, review, and improve agent-based workflows by providing a structured set of patterns, governance rules, and reference material.

Core Features & Use Cases

Core patterns: Plan-First, Prompt Chaining, Routing, Parallelization, Orchestrator-Workers, Evaluator-Optimizer.
Governance: Handoffs, Context Engineering, and IR Architecture to support long-horizon projects.

Quick Start

Review the AGENTS.md and the references folder to align your first workflow.

Frequently Asked Questions about agentic-workflow-guide

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

FAQPage Schema

What are the core patterns for multi-agent workflow orchestration?▼

Multi-agent workflow orchestration relies on core patterns like Plan-First, Prompt Chaining, Routing, Parallelization, Orchestrator-Workers, and Evaluator-Optimizer. These patterns help construct reliable, modular agent ecosystems by enforcing clear I/O contracts and explicit handoffs.

How do I design context engineering for long-horizon agent workflows?▼

Design context engineering for long-horizon agent workflows by applying governance guidance and specific architectural patterns. This approach manages context effectively across extended tasks, ensuring single-source-of-truth (SSOT) and idempotency are maintained throughout the workflow lifecycle.

What is the best way to structure handoffs between sub-agents?▼

The best way to structure handoffs between sub-agents involves enforcing clear I/O contracts and explicit governance rules. This ensures robust multi-agent workflows maintain modularity and reliability during task delegation without losing execution context.

Can I build a single-agent workflow using multi-agent orchestration patterns?▼

Yes, you can build a single-agent workflow using these orchestration patterns. The design principles, including SSOT and idempotency, apply to both single-agent and multi-agent workflows, ensuring reliable execution even without complex sub-agent delegation.

Why do my multi-agent workflows fail on long-horizon tasks?▼

Multi-agent workflows fail on long-horizon tasks due to inadequate context engineering and poorly defined handoffs. Without explicit governance, single-source-of-truth enforcement, and clear I/O contracts, agents lose critical state information during extended execution.