resource-selection

Generate and evaluate 2–4 candidate approaches for sub-tasks.

Updated Mar 14, 2026
One-click install
npx skills add https://github.com/dgroch/metacognition --skill resource-selection
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: resource-selection
Source: https://github.com/dgroch/metacognition/tree/main/skills/resource-selection
Command: npx skills add https://github.com/dgroch/metacognition --skill resource-selection

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deliberate tool selection to avoid defaulting to a single solution when multiple viable approaches exist, reducing waste and errors.

Core Features & Use Cases

  • Deliberate candidate generation: Proactively creates 2–4 viable approaches by consulting capabilities and environment.
  • Evaluation framework: Compares options across fitness, efficiency, reliability, environment fit, reversibility.
  • Commitment protocol: Selects and documents the chosen approach, with recovery paths if a step fails.
  • Coordination with other skills: Integrates with failure recovery, environment model, and task decomposition to maintain coherence.

Quick Start

Generate 2–4 candidate approaches, evaluate them against key criteria, and select the best tool to complete the task.

Frequently Asked Questions about resource-selection

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

FAQPage Schema
How do I pick the optimal tool for each sub-task when multiple approaches exist?

To pick the optimal tool, generate 2–4 candidate approaches by consulting capability inventories and environment constraints, then evaluate them against fitness, efficiency, reliability, and environment fit to select the best option.

What is deliberate tool selection in AI agent decision-making?

Deliberate tool selection is an AI agent decision-making process that proactively generates multiple viable approaches and evaluates them against defined criteria to avoid defaulting to a single solution, reducing waste and errors.

How do I evaluate candidate approaches for a sub-task?

Evaluate candidate approaches by comparing options across fitness, efficiency, reliability, environment fit, and reversibility, leveraging capability inventories and failure history to guide decisions and safeguard against poor choices.

When do I need deliberate resource selection for my environment?

You need deliberate resource selection when multiple viable approaches exist for a sub-task and you must avoid defaulting to a single solution, ensuring the chosen tool aligns with environment constraints and failure history.

Does resource selection work with task decomposition and failure recovery?

Resource selection coordinates with failure recovery, environment model, and task decomposition skills to maintain coherence, providing a commitment protocol with recovery paths if a step fails.

What are the limitations of defaulting to a single tool for sub-tasks?

Defaulting to a single tool for sub-tasks limits your ability to adapt to environment constraints, ignores failure history, and increases waste and errors by bypassing an evaluation framework that checks fitness, efficiency, and reliability.