model-selection

Select cost-effective subagent models (Haiku, Sonnet, Opus) for tasks.

Updated Jun 23, 2025
One-click install
npx skills add https://github.com/Dayopt/app --skill model-selection-dayopt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-selection
Source: https://github.com/Dayopt/app/tree/main/.claude/skills/model-selection
Command: npx skills add https://github.com/Dayopt/app --skill model-selection-dayopt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Subagentモデル選択の自動化により、Taskツール使用時のコストとパフォーマンスのバランスを取るための最適なモデルを選択します。

Core Features & Use Cases

  • Auto-select Haiku, Sonnet, or Opus based on task requirements to optimize cost and performance.
  • Provide guidelines for when to deploy each model during multi-subagent workflows.
  • Use Case: Launch parallel subagents for a search or data investigation and automatically allocate the most cost-effective model per subtask.

Quick Start

Automatically select the most cost-effective subagent model (Haiku, Sonnet, Opus) based on task requirements.

Frequently Asked Questions about model-selection

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

FAQPage Schema
How do I select the most cost-effective subagent model for a task?

Subagent model selection identifies the optimal Claude model (Haiku, Sonnet, or Opus) based on task requirements to balance cost and performance in multi-subagent workflows.

When do I need to use Haiku, Sonnet, or Opus for parallel subagents?

Use Haiku, Sonnet, or Opus for parallel subagents when allocating resources across search or data investigation subtasks that have varying cost constraints and performance requirements.

Can I automate model selection across multi-subagent workflows with cost constraints?

Yes, you can automate model selection across multi-subagent workflows by applying predefined guidelines and exemplars that map subagent types to optimal models based on cost constraints and context availability.

What is the best way to optimize Task tool costs when launching multiple subagents?

The best way to optimize Task tool costs is by automatically allocating the cheapest suitable model per subtask, ensuring alignment with predefined subagent type and model mapping guidelines.

Does subagent model selection work without predefined guidelines for subagent type mappings?

Subagent model selection relies on predefined guidelines and exemplars for subagent type and model mappings to ensure proper alignment with cost constraints and performance requirements.

Why does deploying Opus for every subagent task increase costs unnecessarily?

Deploying Opus for every subagent task increases costs unnecessarily because it ignores cost-effective model selection, which allocates cheaper models like Haiku or Sonnet to suitable low-complexity subtasks.