model-selection

Map development tasks to AI models and effort levels in Teqo.

Updated Jul 31, 2026
One-click install
npx skills add https://github.com/fsolla/teqo --skill model-selection-fsolla
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: model-selection
Source: https://github.com/fsolla/teqo/tree/main/.cursor/skills/model-selection
Command: npx skills add https://github.com/fsolla/teqo --skill model-selection-fsolla

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill eliminates inefficient model usage by providing a standardized framework for selecting the optimal AI model and effort level based on task complexity, ensuring cost-effective and high-quality outcomes.

Core Features & Use Cases

  • Model Mapping: Automatically routes tasks to Composer 2.5, Grok 4.5, or Kimi K3 based on the specific requirements of the issue.
  • Effort Calibration: Provides explicit guidance on selecting low, medium, or high effort levels for Grok-based tasks to match the depth of deliberation required.
  • Bipartite Workflow: Orchestrates complex tasks by splitting them into planning and execution phases, assigning the appropriate model to each stage.

Quick Start

Apply the model-selection skill to determine the correct model and effort level for the current task based on the canonical priority table.

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 optimal AI model for development tasks to balance cost and quality?

Optimize AI model selection by mapping task classes to specific model families and effort levels, balancing deliberation quality with operational costs. This enforces strict model selection rules to prevent inefficient usage and standardize task routing outcomes.

How does effort calibration work when routing tasks to Grok models?

Effort calibration for Grok-based tasks involves selecting low, medium, or high effort levels to match the required depth of deliberation. This aligns computational resources directly with task complexity.

What is the best way to orchestrate complex agentic workflows across multiple AI models?

Orchestrate complex agentic workflows using a bipartite approach that splits tasks into planning and execution phases, assigning the appropriate model to each stage. This ensures optimal deliberation and execution for complex requirements.

Can I use fast-mode variants for AI task routing in Teqo?

No, the model-selection framework enforces strict rules that prevent the use of unauthorized fast-mode variants. It enforces strict model selection rules to ensure outcomes remain high-quality and cost-effective.

How do I route development issues to Composer 2.5, Grok 4.5, or Kimi K3?

Route development issues by mapping their specific requirements to the appropriate model family, such as Composer 2.5, Grok 4.5, or Kimi K3. This automatic routing ensures the selected model fits the task class perfectly.