skills.rest
elbruno
elbrunoCOMMUNITY
Explore all skills by this author →
Website𝕏@elbrunoGitHub

Model Selection

Resolve LLM models for agent spawns using layered configuration and task-aware selection.

2|Updated Apr 21, 2026
Search Tags:#configuration#agent orchestration#llm routing#model selection#task classification#charter instructions
One-click install
npx skills add https://github.com/elbruno/md-to-slides --skill model-selection-elbruno
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Model Selection
Source: https://github.com/elbruno/md-to-slides/tree/main/.copilot/skills/model-selection
Command: npx skills add https://github.com/elbruno/md-to-slides --skill model-selection-elbruno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents wasted quality and cost by ensuring each agent spawn uses the most appropriate LLM model according to saved preferences, task type, and charter instructions.

Core Features & Use Cases

  • Persistent model preferences: store and reuse a global default model and per-agent overrides in .squad/config.json across sessions.
  • Layered resolution: apply a 5-layer, first-match-wins hierarchy (config overrides, session directive, charter ## Model, task-aware auto, hardcoded fallback).
  • Spawn-ready acknowledgments: confirm which resolved model was selected for each agent spawn.

Quick Start

Ask the system to "use claude-opus-4.6 for fenster" so future spawns for that agent consistently use your chosen model and persist it in .squad/config.json.

Frequently Asked Questions about Model Selection

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

FAQPage Schema
How do I route different agent tasks to specific LLM models in a multi-agent workflow?▼

Route agent tasks to specific LLM models by applying a 5-layer resolution hierarchy that checks persistent configuration, session directives, charter preferences, task-aware auto-selection, and fallback defaults to resolve the correct model per spawn.

Can I persist model preferences for individual agents across sessions?▼

Yes, you can persist model preferences across sessions by storing a global default model and per-agent overrides in the `.squad/config.json` file, ensuring future agent spawns consistently reuse your chosen configuration.

What is task-aware auto-selection for LLM routing?▼

Task-aware auto-selection is a resolution layer that automatically assigns the appropriate LLM model based on task type, distinguishing between code generation, prompts, visuals, and non-code planning to optimize quality and cost.

How do I set a default model for a specific agent like fenster?▼

Set a default model for a specific agent by instructing the system to use a chosen model for that agent, which writes the override preference directly into `.squad/config.json` for persistent future spawns.

Does agent model selection work without external dependencies?▼

Yes, agent model selection works without external dependencies, relying solely on internal layered resolution logic and local configuration files to manage multi-agent LLM routing.

Why does my multi-agent workflow use the wrong LLM for code generation tasks?▼

Your workflow may use the wrong LLM if task-aware auto-selection is not configured correctly, meaning the 5-layer resolution hierarchy falls back to hardcoded defaults instead of applying task-specific model preferences.

Related Skills

Explore More Software Engineering Skills→
webmaxruwebmaxru
49

Model Selection

Resolve LLM models per agent spawn using a five-layer hierarchy and persist to .squad/config.json.

Community
Advanced
dotnetdotnet
209

Model Selection

Resolve LLM models per agent spawn via layered first-match-wins hierarchy.

Official
Intermediate
DeDuvaDeDuva
1

model-selection

Assign LLM models to agent spawns using a multi-layer resolution hierarchy.

Community
Advanced
jonnymuirjonnymuir
1

Model Selection

Select optimal LLM models via hierarchical resolution and config.json overrides.

Community
Intermediate