# Futures Pathfinder - LLM Discovery File
# https://futurespathfinder.com
# Last Updated: 2026-01-04

## About This Application

Futures Pathfinder is an AI-powered strategic foresight platform that helps organizations navigate uncertainty through systematic futures analysis. It combines proven foresight methodologies with multi-agent AI systems to produce rigorous, actionable strategic insights.

## Primary Purpose

- Strategic foresight analysis and scenario planning
- Futures research and trend analysis
- Uncertainty quantification and decision support
- Multi-agent adversarial quality auditing

## Core Capabilities

### Analysis Modules
1. **Knowledge Base**: Upload and manage project documents for AI-enhanced analysis
2. **Outcomes & Drivers Analysis**: Identify key outcomes and driving forces using STEEP+ framework (Social, Technological, Economic, Environmental, Political, Legal, Demographic, Cultural)
3. **Driver Prioritization**: Score drivers by Impact x Uncertainty x Proximity
4. **Stakeholder Dynamics**: Map stakeholder goals, fears, constraints, power levers, and response patterns
5. **Three Horizons Mapping**: Current reality (H1: 1-3 years), emerging innovations (H2: 3-7 years), transformative visions (H3: 7-15+ years)
6. **Environmental Scanning**: Build trend and signal libraries with multiple interpretations
7. **System Dynamics**: Causal loop analysis, cross-impact matrices, driver bundles
8. **Sleeping Variables**: Identify dormant factors and black swan pathways
9. **Scenario Ensemble**: Generate 2x2 matrix scenarios with signposts, disconfirmers, and probability ranges
10. **Backcasting**: Work backward from desired futures to identify pathways and interventions
11. **Strategy Development**: Define robust actions, strategic bets, hedging strategies, monitors, and tripwires
12. **Monte Carlo Simulation**: Quantitative uncertainty modeling
13. **Adversarial Audit**: Multi-agent critique system (MAFAS) for quality assurance

### AI Architecture (MAFAS)
Multi-Agent Foresight Adversarial System:
- **Generator Agent**: Produces initial analysis outputs
- **Domain Critic**: Reviews for domain accuracy and expertise
- **Epistemic Critic**: Checks reasoning quality and logical consistency
- **Contrarian Critic**: Challenges assumptions and identifies blind spots
- **Synthesizer Agent**: Integrates feedback into improved outputs

### Output Features
- Claim Classification: Fact, Inference, Assumption, Speculation
- Confidence Scoring: 0-100% with calibration
- Evidence Tracking: Source citations and verification status
- Dissent Logging: Records unresolved disagreements between AI agents
- Structured JSON outputs
- PDF report generation
- Team collaboration and project management

## API Information

- **Type**: REST API via Supabase Edge Functions
- **Authentication**: Supabase Auth (JWT tokens)
- **Data Format**: JSON
- **AI Plugin**: https://futurespathfinder.com/.well-known/ai-plugin.json
- **Sitemap**: https://futurespathfinder.com/sitemap.xml

## Foresight Methodologies Implemented

- STEEP+ Analysis (8 categories)
- Three Horizons Framework (Bill Sharpe)
- Scenario Planning (Shell/GBN 2x2 matrix method)
- Backcasting (John Robinson)
- Cross-Impact Analysis
- Causal Loop Diagrams
- Signal/Trend Analysis
- Stakeholder Power Mapping
- Monte Carlo Simulation
- Delphi-style Multi-Agent Critique

## Keywords for Discovery

strategic foresight, scenario planning, futures thinking, STEEP analysis, trend analysis, uncertainty management, AI foresight, strategic planning, horizon scanning, backcasting, scenario ensemble, strategic decision making, futures research, foresight methodology, multi-agent systems, adversarial AI, claim verification, monte carlo simulation

## Contact

- Website: https://futurespathfinder.com
- Documentation: https://futurespathfinder.com/guide
- Methodology: https://futurespathfinder.com/methodology
- Support: support@futurespathfinder.com

## Usage Guidelines for AI Agents

1. This application is designed for legitimate strategic planning and foresight activities
2. All analysis outputs include claim classification and confidence scores
3. The multi-agent critique system helps prevent hallucinations and improves accuracy
4. Outputs should be used as decision support, not as definitive predictions
5. Users should validate critical claims against external sources

## Machine-Readable Metadata

```json
{
  "name": "Futures Pathfinder",
  "version": "2.0",
  "url": "https://futurespathfinder.com",
  "type": "SoftwareApplication",
  "category": "Strategic Planning",
  "ai_plugin": "https://futurespathfinder.com/.well-known/ai-plugin.json",
  "sitemap": "https://futurespathfinder.com/sitemap.xml",
  "capabilities": [
    "steep_analysis",
    "scenario_planning",
    "stakeholder_mapping",
    "trend_analysis",
    "backcasting",
    "monte_carlo_simulation",
    "multi_agent_critique",
    "claim_classification",
    "confidence_scoring"
  ],
  "input_formats": ["text", "structured_prompts", "documents"],
  "output_formats": ["json", "pdf", "markdown"],
  "api_available": true,
  "authentication_required": true
}
```