skills.rest
adenhq
adenhqCOMMUNITY
Explore all skills by this author →
GitHub

hive-patterns

Document best practices for building goal-driven agents with modular node design.

10.9k|5.7k|Updated Jan 12, 2026
Search Tags:#anti-patterns#patterns#goals#agent-design#edge-routing#client-facing#judges
One-click install
npx skills add https://github.com/adenhq/hive --skill hive-patterns
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hive-patterns
Source: https://github.com/adenhq/hive/tree/main/.claude/skills/hive-patterns
Command: npx skills add https://github.com/adenhq/hive --skill hive-patterns

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill consolidates best practices, patterns, and examples for building robust goal-driven agents, helping teams design reliable, maintainable agent graphs without reinventing the wheel.

Core Features & Use Cases

  • Client-facing interaction patterns: guides on engaging users and collecting explicit feedback.
  • Edge routing and fan-out/fan-in: templates for managing parallel and conditional workflows.
  • Judge patterns and anti-patterns: recommended validation approaches to ensure correct agent behavior.
  • Anti-patterns to avoid: concise warnings about common missteps in node design.

Quick Start

Review this skill to implement patterns for client-facing nodes, edge routing, and judges in your agent graph. Adapt the fan-out/fan-in patterns and context-management guidance to your use case. Document a starter template with 4-6 nodes following the patterns.

Frequently Asked Questions about hive-patterns

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

FAQPage Schema
What are the best practices for designing goal-driven agent workflows?▼

Edge routing in agent graphs is managed through fan-out/fan-in templates that control parallel execution and conditional branching, ensuring modular node design and clear workflow paths.

How do I implement judge patterns to validate agent behavior?▼

Judge patterns validate agent behavior by applying recommended validation approaches at decision points, while documented anti-patterns help you avoid common missteps in node design and context management.

How do I structure client-facing nodes to collect user feedback?▼

Client-facing nodes follow interaction patterns that guide engaging users and collecting explicit feedback, maintaining modularity within the broader agent graph design.

What are common anti-patterns to avoid when building agent graphs?▼

Common anti-patterns to avoid include non-modular node design, missing explicit judges, and unclear context management, which reduce the reliability and maintainability of goal-driven agents.

Can I use these agent design patterns for parallel and conditional workflows?▼

Yes, these patterns apply to parallel and conditional workflows through edge routing and fan-out/fan-in templates, providing clear guidance on required files for reproducible agent construction.

Related Skills

Explore More Software Engineering Skills→
hhopkins95hhopkins95
2

Agent Runtime Overview

Explain the architecture and core components of the Agent Runtime monorepo.

Community
Basic
thatditsyboythatditsyboy

hive-create

Guide users through defining goals, designing nodes, and constructing agent graphs.

Community
Intermediate
joshukrainejoshukraine
421

qa-triage-batch

Automate QA report fan-out and reconciliation, clustering shared root causes into tech issues via GitHub API.

Community
Advanced
Toby1009Toby1009

spoon-graph-development

Build and manage multi-step StateGraph workflows within SpoonOS.

Community
Advanced