mcpField
W
WorldThreatModelHarness

WorldThreatModelHarness

github.com/danielmiesslerProductivity19.6k104 installsUpdated today
Repository

About WorldThreatModelHarness

WorldThreatModelHarness is a skill for testing ideas, strategies, and investments against world models spanning multiple time horizons. It supports different analysis depths and combines adversarial analysis with first-principles work and multi-perspective debate.

Key features

  1. 01Tests inputs across 11 time horizons
  2. 02Offers fast, standard, and deep tiers
  3. 03Uses RedTeam for adversarial analysis
  4. 04Integrates FirstPrinciples and Council
  5. 05Supports updating and viewing world models

Use cases

  1. 01Stress-test a strategy across time horizons
  2. 02Assess investments against future scenarios
  3. 03Conduct deep analysis for high-stakes decisions

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/danielmiessler/lifeos --skill worldthreatmodelharness

After installing, type/worldthreatmodelharnessin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

OutputFormat.md
ModelTemplate.md
Workflows/
TestIdea.md
ViewModels.md
UpdateModels.md

Details

From the project README

World Threat Model Harness

A system of 11 persistent world models spanning 6 months to 50 years. Each model is a deep (~10 page) analysis of geopolitics, technology, economics, society, environment, security, and wildcards for that time horizon. Ideas, strategies, and investments are tested against ALL horizons simultaneously using adversarial analysis (RedTeam, FirstPrinciples, Council).

Workflow Routing

TriggerWorkflowDescription
"test idea", "test strategy", "test investment", "how will this hold up", "stress test", "test against future"Workflows/TestIdea.mdTest any input against all 11 world models
"update world model", "update models", "refresh models", "new analysis"Workflows/UpdateModels.mdRefresh world model content with new research/analysis
"view world model", "show models", "current models", "model status"Workflows/ViewModels.mdRead and summarize current world model state

Tier System

All workflows support three execution tiers:

TierTarget TimeStrategyWhen to Use
Fast~2 minSingle agent synthesizes across all modelsQuick gut-check, casual exploration
Standard~10 min11 parallel agents + RedTeam + FirstPrinciplesMost use cases, good depth/speed balance
DeepUp to 1 hr11 parallel agents + per-horizon Research + RedTeam + Council + FirstPrinciplesHigh-stakes decisions, major investments

Default tier: Standard. User specifies with "fast", "deep", or tier defaults to Standard.

World Model Storage

Models are stored at: $PAI_DIR/MEMORY/RESEARCH/WorldModels/

FileHorizon
INDEX.mdSummary of all models with last-updated dates
6-month.md6-month outlook
1-year.md1-year outlook
2-year.md2-year outlook
3-year.md3-year outlook
5-year.md5-year outlook
7-year.md7-year outlook
10-year.md10-year outlook
15-year.md15-year outlook
20-year.md20-year outlook
30-year.md30-year outlook
50-year.md50-year outlook

Context Files

FilePurpose
ModelTemplate.mdTemplate structure for world model documents
OutputFormat.mdTemplate for TestIdea results output

Skill Integrations

This skill orchestrates multiple PAI capabilities:

  • RedTeam — Adversarial stress testing of ideas against each horizon
  • FirstPrinciples — Decompose idea assumptions into hard/soft/assumption constraints
  • Council — Multi-perspective debate on idea viability across horizons
  • Research — Deep research for model creation and updates

Voice Notification

Before any workflow execution:

curl -s -X POST http://localhost:8888/notify \
  -H "Content-Type: application/json" \
  -d '{"message": "Running WORKFLOW_NAME in the World Threat Model Harness", "voice_id": "fTtv3eikoepIosk8dTZ5"}'

Customization Check

Before execution, check for user customizations at: ~/.claude/PAI/USER/SKILLCUSTOMIZATIONS/WorldThreatModelHarness/

Comments

Markdown supported