skills.rest
codewithmukesh
codewithmukeshOFFICIAL
Explore all skills by this author →
Website𝕏@iammukeshmGitHub

model-selection

Recommend Claude models for .NET development tasks with cost-aware routing.

640|145|Updated Feb 20, 2026
Search Tags:#claude#subagents#cost-optimization#opus#model-selection#haiku#sonnet
One-click install
npx skills add https://github.com/codewithmukesh/dotnet-claude-kit --skill model-selection-codewithmukesh
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: model-selection
Source: https://github.com/codewithmukesh/dotnet-claude-kit/tree/main/skills/model-selection
Command: npx skills add https://github.com/codewithmukesh/dotnet-claude-kit --skill model-selection-codewithmukesh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps .NET developers and Claude orchestrators pick the most cost-effective and capability-appropriate Claude model (Opus, Sonnet, or Haiku) for a given task so that reasoning-heavy work uses stronger models and routine tasks use faster, cheaper models.

Core Features & Use Cases

  • Model-to-task mapping: Clear guidance on when to use Opus 4.6 for deep reasoning, Sonnet 4.6 for high-throughput implementation, and Haiku 4.5 for lightweight subagent work.
  • Phase-based workflow: Supports Plan (Opus), Execute (Sonnet), Review (Opus) patterns for end-to-end feature delivery.
  • Subagent routing & cost control: Assigns models to subagents by complexity and optimizes for context-window budgeting and cost efficiency.
  • Anti-patterns and decision tables: Practical cautions (e.g., don't use Opus for routine tasks) and a scenario-to-model decision guide.

Quick Start

Use the model-selection skill to evaluate a task and recommend Opus, Sonnet, or Haiku with a one-line rationale and recommended routing.

Frequently Asked Questions about model-selection

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

FAQPage Schema
How do I choose the right Claude model for .NET development tasks?▼

Choosing the right Claude model for .NET development involves matching model capability to task complexity: Opus 4.6 for deep reasoning, Sonnet 4.6 for high-throughput implementation, and Haiku 4.5 for lightweight subagent work.

When should I use Claude Opus vs Sonnet for coding workflows?▼

Use Claude Opus for reasoning-heavy phases like planning and code review, and use Sonnet for high-throughput implementation tasks. This phase-based workflow ensures deep reasoning uses stronger models while routine work uses faster, cheaper models.

How do I optimize Claude API costs when using subagents?▼

Optimize Claude API costs with subagents by assigning Haiku 4.5 to lightweight tasks and reserving Opus 4.6 for complex reasoning, applying context-window budgeting and cost-aware routing to minimize expenses.

What are common anti-patterns when assigning Claude models to tasks?▼

A common anti-pattern when assigning Claude models is using Opus for routine tasks, which wastes resources. Follow decision tables to route work by complexity, ensuring fast, cheaper models handle lightweight subagent operations.

Can I apply phase-based model routing to .NET codebases?▼

Yes, you can apply phase-based model routing to .NET codebases using a Plan (Opus), Execute (Sonnet), Review (Opus) pattern for end-to-end feature delivery, ensuring capability-appropriate model assignment across all phases.

How does context-window budgeting affect subagent model selection?▼

Context-window budgeting affects subagent model selection by constraining the context size per subagent, requiring cost-aware routing that assigns Haiku for lightweight work and Opus only when deep reasoning is explicitly needed.

Related Skills

Explore More Software Engineering Skills→
ngocsangyemngocsangyem
14

meow:scale-routing

Classify task descriptions into domain, complexity, workflow, and model tier.

Community
Advanced
campbellmcgregorcampbellmcgregor

agentic-development

Guide AI-assisted software builds with workflows, model selection, and prompt strategies.

Community
Basic
novatech2210-cmdnovatech2210-cmd

gsd-execute-phase

Execute phase plans in parallel using wave-based scheduling with subagents.

Community
Advanced
planetaryescapeplanetaryescape
7

auto-router-patterns

Route user messages to optimal AI models via task classification and cost-aware tier selection.

Official
Advanced