agent-patterns

Load agent engineering patterns from a personal wiki for multi-agent system design.

Updated Sep 26, 2025
One-click install
npx skills add https://github.com/vietbui1999ru/dotfiles --skill agent-patterns-vietbui1999ru
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-patterns
Source: https://github.com/vietbui1999ru/dotfiles/tree/main/llm-wiki-plugin/skills/agent-patterns
Command: npx skills add https://github.com/vietbui1999ru/dotfiles --skill agent-patterns-vietbui1999ru

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you avoid ad-hoc agent design by loading established agent engineering patterns from your personal wiki before you architect or review multi-agent systems.

Core Features & Use Cases

  • Pattern loading for agent architecture: Guides decisions about when to use a skill versus a subagent versus a team, and how to structure multi-agent workflows.
  • Context management guidance: Supports designing for context filling, degradation, and compression in long-running interactions.
  • Harness, loop, and tool design: Provides approaches for long-horizon harness architecture, iterative loop strategies, and tool design for agents, including verification and security considerations.

Quick Start

Ask your assistant to load the agent engineering patterns from the wiki and then use them to propose an architecture for your multi-agent workflow.

Frequently Asked Questions about agent-patterns

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

FAQPage Schema
How do I design multi-agent workflows using established patterns?

Design multi-agent workflows by loading agent engineering patterns from your wiki to guide architecture decisions. This prevents ad-hoc design by providing established concepts for structuring skills, subagents, and teams within your system.

What is context management strategy for long-running agent interactions?

Context management strategy for long-running agent interactions involves planning for context filling, degradation, and compression. Applying engineering patterns helps you structure how agents maintain and process information over extended multi-step tasks.

How do I set up a verification pipeline and security controls for agent tools?

Set up a verification pipeline and security controls for agent tools by loading specific engineering patterns from your wiki. These patterns provide approaches for iterative loop strategies, tool design, and necessary security considerations for your architecture.

When do I need a long-horizon harness architecture for multi-agent systems?

You need a long-horizon harness architecture for multi-agent systems when designing extended iterative loops and complex tool interactions. Loading engineering patterns helps you determine the appropriate harness structure and loop strategies for your tasks.

Can I use my personal wiki for agent architecture pattern retrieval?

Yes, you can use your personal wiki for agent architecture pattern retrieval. The system performs a wiki-page search using configured query modalities and scoring thresholds to return relevant concept pages for your requested task types.

What is the best way to choose between using a skill, subagent, or team?

The best way to choose between a skill, subagent, or team is to load engineering patterns from your wiki. These patterns guide your architectural decisions on when to use each component type to effectively structure your multi-agent workflows.