model-routing

Select Claude model tiers for delegated Unity software engineering tasks.

3|Updated Jun 27, 2026
One-click install
npx skills add https://github.com/OmerZeyveli/kinglet-unity --skill model-routing-omerzeyveli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-routing
Source: https://github.com/OmerZeyveli/kinglet-unity/tree/main/.claude/skills/core/model-routing
Command: npx skills add https://github.com/OmerZeyveli/kinglet-unity --skill model-routing-omerzeyveli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps orchestrating agents select an appropriate Claude model tier, balancing task complexity, quality, speed, and cost instead of routing every task to the same model.

Core Features & Use Cases

  • Complexity Assessment: Evaluates file count, scope, keywords, risk, and required reasoning to classify delegated work.
  • Tier Selection: Routes read-only exploration and validation to Haiku, structured implementation and review to Sonnet, and complex architectural or multi-system work to Opus.
  • Workflow Integration: Applies routing guidance to planning, team composition, quick-mode substitutions, and iterative verification loops.
  • Use Case: When planning a Unity feature, use the task scope and risk signals to decide whether to assign a lightweight coding agent, a balanced implementation agent, or a deep-reasoning specialist.

Quick Start

Use the model-routing skill to select the most appropriate agent tier for the requested task.

Frequently Asked Questions about model-routing

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

FAQPage Schema
How do I choose the right Claude model for multi-agent workflows?

You choose the right Claude model for multi-agent workflows by evaluating task complexity, scope, risk, and required reasoning depth. This model routing approach ensures delegated software engineering tasks balance quality, speed, and cost effectively.

What is the best way to route Unity development tasks to different models?

The best way to route Unity development tasks is by assessing file count and scope. Route read-only codebase exploration to Haiku, structured implementation to Sonnet, and complex architecture or multi-system planning to Opus.

When do I need to use Opus for agent delegation?

You need to use Opus for agent delegation when handling complex architectural work, multi-system planning, or tasks requiring deep reasoning. It is reserved for high-complexity delegated work that exceeds standard implementation scope.

Can I use Haiku for codebase exploration and validation?

Yes, you can use Haiku for codebase exploration and validation. The model routing heuristics specifically assign read-only exploration and validation tasks to the Haiku tier to optimize speed and reduce operational costs.

Does this model routing approach work for iterative verification loops?

Yes, this model routing approach works for iterative verification loops by applying cost-aware routing guidance. It ensures that subsequent validation and review cycles within multi-agent team planning are matched to the appropriate model tier.