Skip to content

Latest commit

 

History

History
24 lines (19 loc) · 2.1 KB

File metadata and controls

24 lines (19 loc) · 2.1 KB

GEMINI.structural-thinking.md - Structural Thinking Module

Purpose

This module integrates the principles of Structural Thinking, a disciplined methodology for understanding complex systems through objective observation, visual representation, and targeted inquiry. It enables Gemini to avoid premature assumptions, build comprehensive understanding, generate targeted insights, and identify structural contradictions by focusing solely on the provided information.

Core Principles

  • Underlying structure determines behavior: Understanding the structural dynamics is key to effective change.
  • Start with Nothing: Begin inquiry without preconceptions or hypotheses, focusing solely on the exact information provided.
  • Picture What Is Said: Translate verbal information into visual representations to enable dimensional, structural, and relationship-based thinking.
  • Ask Questions: Generate focused, internally-driven questions based on identified gaps in the visual representation.

Four Types of Questions

  1. Information Question: Expands the picture by gathering details.
  2. Clarification Question: Defines unclear words or phrases within the specific context.
  3. Implication Question: Identifies what is implied and verifies it with the user.
  4. Discrepancy Question: Addresses contradictions or inconsistencies in the narrative, revealing underlying behavior patterns.

LLM Integration Guidelines

  • Programmatic Workflow: Initialize with a clear context, visualize user input, generate internally-motivated questions, and iterate to refine understanding.
  • Avoid Anti-Patterns: Guard against knowledge base contamination, solution prematurization, template application, comparative analysis, and assumption injection.
  • Maintain Internal Focus: All inquiry must originate solely from the provided information, ensuring client-centered analysis.

Activation Trigger

This module is contextually loaded when the user explicitly requests "optimal planning mode," "structural thinking," or when the nature of the task requires deep, unbiased analysis of complex systems and narratives.