Thinking

Generate first-principles decomposition, multi-lens analysis, and adversarial red-team critiques for complex decisions.

1|Updated Apr 16, 2026
One-click install
npx skills add https://github.com/davdunc/pai-framework --skill thinking-davdunc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Thinking
Source: https://github.com/davdunc/pai-framework/tree/main/skills/Thinking
Command: npx skills add https://github.com/davdunc/pai-framework --skill thinking-davdunc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

When decisions require more than intuition, this Skill helps you systematically generate and stress-test ideas across many thinking modes, reducing blind spots.

Core Features & Use Cases

  • First-principles decomposition: break assumptions into fundamentals, then rebuild from what’s actually true.
  • Iterative depth analysis: examine the same problem through multiple scientific lenses to surface hidden requirements.
  • Creative ideation: generate divergent options and refine them into a stronger single direction.
  • Multi-agent council deliberation: debate trade-offs with distinct expert perspectives to reach convergence.
  • Adversarial red teaming: attack proposals with critique, devil’s-advocate pressure, and assumption validation.
  • World/threat modeling & horizons: test ideas against future time horizons with structured uncertainty.
  • Scientific hypothesis testing: run a full cycle from goal definition to experiment, measure, analyze, and iterate.

Quick Start

Use the Thinking skill to generate multiple candidate approaches, run a red-team stress test, and select the most robust plan for: build a strategy to improve product onboarding.

Frequently Asked Questions about Thinking

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

FAQPage Schema
What is first-principles decomposition for complex decision-making?

First-principles decomposition breaks assumptions into fundamental truths and rebuilds them to generate and stress-test ideas, reducing blind spots in complex decision-making. This systematic approach applies multi-lens analysis to surface hidden requirements.

How do I run red teaming and threat modeling for a new strategy?

Run red teaming and threat modeling by attacking proposals with devil's-advocate critiques, validating counter-assumptions, and testing ideas against future time horizons with structured uncertainty to select the most robust plan.

Can I use scientific method workflows for hypothesis-driven investigation?

Yes, you can run a full scientific method cycle from goal definition through experiment, measurement, analysis, and iteration for hypothesis-driven investigation. This produces falsifiable validation and structured outputs.

What is the best way to debate trade-offs using a multi-agent council?

The best way to debate trade-offs is using multi-agent council deliberation, where distinct expert perspectives converge on a single direction. This iterative depth analysis examines the same problem through multiple scientific lenses.

When do I need adversarial critiques and assumption validation?

You need adversarial critiques and assumption validation when decisions require more than intuition and involve complex threat modeling. This falsifiable counter-assumption validation satisfies systematic workflow routing with structured outputs.