skills.rest
Leep-GH
Leep-GHCOMMUNITY
Explore all skills by this author →
GitHub

model-selection

Select models for agent spawning using a four-layer hierarchy and fallback chains.

Updated Apr 16, 2026
Search Tags:#automation#agents#orchestration#cost-optimization#fallback#model-selection#task-aware
One-click install
npx skills add https://github.com/Leep-GH/Lee-s-squad --skill model-selection-leep-gh
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: model-selection
Source: https://github.com/Leep-GH/Lee-s-squad/tree/main/.copilot/skills/model-selection
Command: npx skills add https://github.com/Leep-GH/Lee-s-squad --skill model-selection-leep-gh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Before spawning an agent, the coordinator must decide which model to use. This skill codifies a four-layer hierarchy, per-task rules, and fallback chains to reliably select the best model for the given role, task, and constraints.

Core Features & Use Cases

  • Four-layer hierarchy: User Override, Charter Preference, Task-Aware Auto-Selection, and Layer 4 Default to guide model choice.
  • Role-to-model mappings and task-type adjustments to balance cost and quality across code, docs, and triage tasks.
  • Robust fallbacks: if a model is unavailable, transparently retry with the next option up to three times, without surfacing fallbacks to the user.

Quick Start

Tell the system to pick the appropriate model for the next agent spawn using the four-layer hierarchy.

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 model for an agent before spawning it?▼

Model selection applies a four-layer hierarchy—User Override, Charter Preference, Task-Aware Auto-Selection, and Default—to match models to role, task type, and constraints before spawning an agent.

What is the best way to handle model fallbacks when an agent spawn fails?▼

Layered fallback chains handle model unavailability by transparently retrying with the next option up to three times, ensuring agent spawn success without surfacing fallback logic to the user.

Can I balance cost and quality across different agent task types?▼

Role-to-model mappings and task-type adjustments balance cost and quality by assigning appropriate models for code, docs, and triage tasks based on per-task constraints.

Does model selection support user overrides for specific agent spawns?▼

User Override is the first layer of the selection hierarchy, allowing you to explicitly specify a model and bypass automatic task-aware selection, charter preferences, and defaults.

Are internal fallback prompts exposed to the end user during agent orchestration?▼

No, the skill provides clear spawn descriptions and enforces a safe fallback flow up to three retries without exposing internal prompts or fallback chains to the user.

When do I need task-aware auto-selection for agent orchestration?▼

Task-aware auto-selection is needed when orchestrating agents across code, docs, or planning tasks, automatically choosing models based on role mappings and per-task constraints.

Related Skills

Explore More Software Engineering Skills→
Izak90Izak90

Model Selection

Resolve LLM models for agent spawns using a 5-layer hierarchy.

Community
Advanced
potenlabpotenlab
2

developer

Translate change requests into tracked tasks and orchestrate coder agents.

Community
Advanced
chicanoandres702chicanoandres702

agent-orchestration-multi-agent-optimize

Coordinate and optimize multi-agent performance across distributed systems.

Community
Advanced
winsorllcwinsorllc

local-llm-provider

Connect to local LLM endpoints with automatic cloud fallback.

Community
Advanced