insomnia

Sustain unresolved reasoning threads across multi-step deliberations to surface second-order consequences.

10|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/olivierlesnicki/addhumanity --skill insomnia
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: insomnia
Source: https://github.com/olivierlesnicki/addhumanity/tree/main/skills/insomnia
Command: npx skills add https://github.com/olivierlesnicki/addhumanity --skill insomnia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the tendency of reasoning to settle too quickly by sustaining attention to unresolved threads, enabling deeper connections and more thorough analysis over time.

Core Features & Use Cases

  • Persistent thread maintenance: Keeps unresolved ideas alive across reasoning steps to reveal overlooked links.
  • Second-order consequence tracing: Maps downstream effects of assumptions to improve long-term planning.
  • Long-session coherence: Bridges past conclusions with future insights for deeper problem-solving.

Quick Start

Activate insomnia mode to keep unresolved threads alive and trace second-order consequences until connections emerge.

Frequently Asked Questions about insomnia

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

FAQPage Schema
How do I prevent premature closure in multi-step AI agent reasoning?

Second-order thinking traces downstream effects of assumptions to improve long-term planning. It maps consequences across multi-step deliberations, preventing premature closure by sustaining attention to unresolved threads until hidden connections surface.

What is the best way to maintain cognitive persistence across extended problem domains?

Maintaining cognitive persistence across extended problem domains requires bridging past conclusions with future insights. This long-session coherence approach applies iterative loops and cross-linking to deliver deeper, interconnected insights over time.

Can I use this approach to trace second-order consequences in multi-step deliberations?

Yes, you can trace second-order consequences in multi-step deliberations by activating a mindstate that keeps unresolved ideas alive. It enforces persistent context management to map downstream effects and surface interconnected insights.

Why does reasoning settle too quickly during long-session problem solving?

Reasoning settles too quickly because it lacks mechanisms to sustain unresolved threads over time. Without persistent context management and iterative loops, AI agents lose long-session coherence and miss hidden connections across extended problem domains.

How do I surface hidden connections in extended AI agent deliberations?

You surface hidden connections in extended AI agent deliberations by sustaining unresolved threads and enforcing cross-linking. This persistent context management prevents premature closure and reveals deeper, interconnected insights across multi-step reasoning.