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

model-selection

Resolves which LLM model to assign for each agent spawn using a layered configuration hierarchy.

2|Updated Jul 24, 2026
Search Tags:#multi-agent#orchestration#config#llm routing#model selection#fallback#agent spawn
One-click install
npx skills add https://github.com/elbruno/ElBruno.MagenticUI --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/ElBruno.MagenticUI/tree/main/.squad/templates/skills/model-selection
Command: npx skills add https://github.com/elbruno/ElBruno.MagenticUI --skill model-selection-elbruno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Multi-agent orchestration requires choosing the right LLM for every agent spawn, balancing cost, speed, and capability. Without a consistent resolution strategy, model choices become ad hoc, user preferences get lost between sessions, and spawn acknowledgments lack transparency about which model is running. ## Core Features & Use Cases - 5-Layer Resolution Hierarchy: Resolves models in first-match-wins order across persistent config overrides, session directives, charter preferences, task-aware auto-selection, and a hardcoded fallback. - Persistent Preferences: Saves defaultModel and per-agent agentModelOverrides to .squad/config.json so choices survive across sessions. - Task-Aware Auto-Selection: Maps task types to tiers, sending code work to sonnet, visual analysis to opus, and docs or triage to haiku. - Tier-Constrained Fallback Chains: Retries within the same tier when a model is unavailable, never falling up from fast to premium. - Use Case: A user says "always use opus for Fenster"; the skill validates the model ID, writes the override to config.json, and every future spawn of that agent acknowledges the resolved model. ## Quick Start Ask the coordinator to spawn an agent for a coding task and confirm which model was resolved in the spawn acknowledgment.

Frequently Asked Questions about model-selection

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

FAQPage Schema
How do I set a default model for all agent spawns?▼

Say a trigger phrase like "always use X" or "default to X" and the skill validates the model ID against the catalog, then writes defaultModel to .squad/config.json. The preference persists across sessions until cleared.

How do I assign a specific model to one agent?▼

Use a per-agent trigger such as "use X for {agent}". The skill writes the model to agentModelOverrides.{agent} in .squad/config.json, which takes priority over the global defaultModel for that agent.

Which model is selected automatically for coding tasks?▼

Task-aware auto-selection routes code work such as implementation, tests, refactoring, and bug fixes to claude-sonnet-4.6. Non-code tasks like docs, planning, and triage go to claude-haiku-4.5, and visual analysis goes to claude-opus-4.6.

What happens when a selected model is unavailable?▼

The skill retries within the same tier using defined fallback chains, for example premium falls from claude-opus-4.6 to claude-opus-4.5 then claude-sonnet-4.6. It never falls up in tier, so a fast task never lands on a premium model.

How do I clear a saved model preference?▼

Say "switch back to automatic", "clear model preference", or "use default models". The skill removes defaultModel from .squad/config.json and returns to automatic layered selection.

Does a persistent config preference override task-based selection?▼

Yes. Layer 0 config values beat everything else in the first-match-wins hierarchy. If defaultModel is saved, every agent gets that model regardless of role or task type until the preference is cleared.

Related Skills

Explore More Software Engineering Skills→
Izak90Izak90

Model Selection

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

Community
Advanced
Xza85hrfXza85hrf

cost-aware-llm-pipeline

Route tasks across Opus, Sonnet, Haiku, Ollama Cloud, and Ollama Local to minimize token costs.

Community
Intermediate
salvo-rssalvo-rs
20

salvo-proxy

Forward client requests to configurable backend services via a reverse proxy.

Official
Advanced
ZaxbyHubZaxbyHub

post-removal-sweep

Verifies complete removal of code patterns through multi-variant grep searches across codebases.

Official
Intermediate