WorldThreatModel

How This Skill Approaches It

Eleven persistent world models, one per time horizon from 6 months to 50 years, each stored at MEMORY/RESEARCH/WorldModels/. Every model covers geopolitics, technology, economics, society, environment, security, and wildcards for that specific window. When you run TestIdea, the idea gets tested against all 11 simultaneously and returns a probability-weighted scenario matrix — not a flat list. Three execution tiers let you pick depth: Fast runs a single synthesizer agent for a quick check, Standard spins up 11 parallel agents plus RedTeam and FirstPrinciples for most decisions, Deep adds per-horizon Research agents plus Council debate for high-stakes calls. Models persist between sessions and are updated when major events happen, so the analysis reflects accumulated world knowledge rather than starting cold each time. UpdateModels refreshes content; ViewModels lets you audit current state before trusting the output.

• ▸ Each horizon model is a deep (~10 page) analysis of geopolitics, technology, economics, society, environment, security, wildcards
• ▸ Three tiers: Fast (~2 min, single synthesizer), Standard (~10 min, 11 parallel agents + RedTeam + FirstPrinciples), Deep (~1hr, adds Research + Council)
• ▸ Orchestrates RedTeam, FirstPrinciples, Council, Research

In Action

What you say to your DA, and what the WorldThreatModel skill actually does.

• You say "threat model my bet on ai-first content creation"
  → Runs TestIdea (Standard tier): 11 parallel agents each analyze the thesis against their horizon model, RedTeam attacks the assumptions, FirstPrinciples decomposes the hard vs soft constraints, and the output is a probability-weighted scenario matrix showing how the bet holds up from 6 months to 50 years.
• You say "deep threat model this investment — it's a major decision"
  → Runs TestIdea at Deep tier: 11 parallel agents plus per-horizon Research agents pulling fresh data, RedTeam adversarial stress test, and Council debate across multiple expert perspectives — up to an hour of analysis for a high-stakes call.
• You say "update the world models, a lot has changed this year"
  → Runs UpdateModels: refreshes each of the 11 horizon files with current research and analysis, updates INDEX.md with new last-updated timestamps, so subsequent TestIdea runs reflect the current state of the world.

Inside the Skill

The thinking, frameworks, and architecture that distinguish this skill from a generic version of the same task.

User: "threat model my bet on AI-first content creation"
→ Analyzes across 11 time horizons (6mo to 50yr)
→ Identifies structural risks at each horizon
→ Returns probability-weighted scenario matrix

User: "what could go wrong with our newsletter business model?"
→ Maps threat vectors: market, technology, regulatory, competitive
→ Returns prioritized risk register with mitigations

Workflows · 3

1. 01
   TestIdea Workflows/TestIdea.md

   test idea, test strategy, test investment, how will this hold up, stress test, test against future — test any input against all 11 world models

2. 02
   UpdateModels Workflows/UpdateModels.md

   update world model, update models, refresh models, new analysis — refresh world model content with new research/analysis

3. 03
   ViewModels Workflows/ViewModels.md

   view world model, show models, current models, model status — read and summarize current world model state

How to Invoke

Say any of these to your DA and PAI activates the WorldThreatModel skill automatically:

• "threat model"
• "world model"
• "test idea"
• "future analysis"
• "time horizon"
• "stress test against future"
• "long-term risk"

Or invoke explicitly:

Skill("WorldThreatModel")

References · 2

Auxiliary files the skill loads at runtime — frameworks, guides, configs.

• ModelTemplate
• OutputFormat

Related Skills