Version 1.0.0 | MIT License | Built for 2026 and Beyond
Strategic Horizon AI is not another forecasting dashboard—it is a cognitive companion for decision-makers who refuse to be blindsided by uncertainty. Inspired by the Institute for the Future's (IFTF) proven methodologies, this engine fuses soft predictive intelligence (powered by Claude's nuanced reasoning) with hard deterministic chains (Python-based causal modeling) to generate rich, actionable scenario landscapes.
Think of it as a time-traveling cartographer: it maps multiple plausible futures, reveals hidden driver relationships, and helps you backcast from a desired outcome to today's first step. Whether you are a corporate strategist, policy analyst, or startup founder, this tool turns ambiguity into architectural clarity.
"The future is not a destination to predict, but a landscape to navigate."
Traditional forecasting tries to hit a single bullseye. Strategic Horizon AI instead illuminates the entire dartboard—showing you where the wind shifts, where the board might tilt, and which throws have the highest probability of success across multiple scenarios.
| Feature | What It Does |
|---|---|
| Hybrid Prediction Engine | Blends Claude's contextual understanding (soft) with Python's causal chain calculations (hard) |
| Cross-Impact Analysis | Reveals how one driver influences another — a cascading domino map of your strategic landscape |
| Four-Scenario Generation | Automatically builds archetypal scenario quadrants (Growth, Collapse, Discipline, Transformation) |
| Backcasting Workbench | Start from a desired future state; trace backward to today's leverage points |
| Responsive Web UI | Fully mobile-optimized with dark mode, accessibility tags, and multilingual support (EN, ES, FR, DE, JA, ZH) |
| 24/7 API Gateway | Deploy as a microservice; integrate into Slack, Teams, or your own dashboard |
| Multilingual Output | Scenarios rendered in your language of choice — ideal for global teams |
| Export Ready | PDF, Markdown, JSON, and interactive HTML exports |
flowchart TD
A[User Input: Domain + Time Horizon] --> B[Driver Library]
B --> C{Soft Predict (Claude API)}
B --> D{Hard Predict (Python Chain)}
C --> E[Qualitative Scenarios]
D --> F[Quantitative Drivers + Cross-Impact Matrices]
E --> G[Scenario Fusion Engine]
F --> G
G --> H[Four Archetypal Futures]
H --> I[Backcasting Path Generator]
I --> J[Interactive Dashboard]
H --> K[Export / Integration]
- Corporate Strategists seeking resilience in volatile markets
- Policy Makers needing evidence-based foresight for 5–10 year plans
- Product Managers aligning roadmaps with shifting user behaviors
- Academic Researchers studying decision-making under uncertainty
- Startup Founders stress-testing business models against plausible futures
Create a profile.yaml file to define your domain, time horizon, and key drivers:
domain: "renewable-energy-storage"
time_horizon_years: 7
language: "en"
drivers:
- name: "battery_cost_per_kwh"
trend: "exponential_decrease"
uncertainty: "medium"
- name: "government_subsidy_level"
trend: "volatile_increase"
uncertainty: "high"
- name: "public_adoption_rate"
trend: "sigmoid_growth"
uncertainty: "low"
soft_predict_model: "claude-3-opus-2026"
output_format: "scenario_quadrant"
backcast_goal: "achieve 80% grid storage from renewables by 2033"python horizon.py --profile my_profile.yaml --export pdf --lang enExpected output:
[STRATEGIC HORIZON AI] Loading profile: renewable-energy-storage
[STRATEGIC HORIZON AI] Running Claude soft-predict... 🧠
[STRATEGIC HORIZON AI] Running cross-impact calculation... 🔄
[STRATEGIC HORIZON AI] Generating four scenarios...
→ Scenario A: Green Boom
→ Scenario B: Subsidy Cliff
→ Scenario C: Tech Plateau
→ Scenario D: Policy Gridlock
[STRATEGIC HORIZON AI] Backcasting from 'Green Boom'...
[STRATEGIC HORIZON AI] Exporting PDF report... ✅
Strategic Horizon AI supports dual-API integration for maximum flexibility:
from strategic_horizon import HorizonEngine
engine = HorizonEngine(
claude_api_key="sk-claude-...",
openai_api_key="sk-openai-..." # optional fallback
)
scenarios = engine.run(
domain="urban_mobility_2040",
horizon_years=15,
method="hybrid"
)Why both?
- Claude (default): Superior for nuanced, context-rich soft predictions
- OpenAI (fallback): Useful when Claude API is rate-limited or when running cost-sensitive batch jobs
| Operating System | Emoji Rendering | Expected Behavior |
|---|---|---|
| macOS 14+ | Full color | All emojis render correctly |
| Windows 11 | Full color | Minor rendering differences with flags |
| Linux (GNOME) | Full color | Requires fonts-noto-color-emoji |
| Android 14+ | Full color | Perfect in Chrome |
| iOS 16+ | Full color | Perfect in Safari |
| ChromeOS 2026 | Full color | Supported natively |
| Terminal (TTY) | Monochrome | Emojis shown as text fallback |
- Python 3.10+
- Claude API key (required)
- Node.js 18+ (for UI)
git clone https://github.com/your-org/strategic-horizon-ai.git
cd strategic-horizon-ai
pip install -r requirements.txt
cd ui && npm install && cd ..Console mode:
python horizon.py --domain "healthcare_ai_2027" --drivers policy,innovation,costWeb UI mode:
cd ui && npm startVisit http://localhost:3000 to access the dashboard.
Feed in your top 5 market uncertainties — the engine will generate a strategic playbook for each scenario.
Simulate how a new regulation cascades through interconnected drivers (economic, environmental, social).
Backcast from a 2030 market leadership position to today's product roadmap.
Model how shifting weather patterns, policy changes, and technological breakthroughs interact over 10 years.
Strategic Horizon AI is a decision-support tool, not a crystal ball.
- All predictions are probabilistic, not deterministic.
- Outputs should be validated by domain experts before strategic decisions.
- The tool's accuracy depends on the quality of input data and driver definitions.
- No AI model can foresee black swan events or geopolitical blackouts.
- Use this tool to expand your thinking, not to replace your judgment.
This project is licensed under the MIT License — you are free to use, modify, and distribute it for commercial or personal projects.
| Layer | Technology |
|---|---|
| Frontend | React 18 + Tailwind CSS + TypeScript |
| Backend | FastAPI (Python 3.10+) |
| AI Models | Claude API (primary) + OpenAI API (fallback) |
| Data Processing | Pandas + NumPy + NetworkX |
| Scenario Generation | Custom IFTF-inspired algorithm |
| Export | ReportLab (PDF), Pandoc (Markdown), Plotly (HTML) |
| Deployment | Docker + AWS / GCP / Azure-compatible |
This repository is optimized for discovery by terms such as:
- Strategic foresight engine
- IFTF scenario planning tool
- Claude AI predictions
- Cross-impact analysis software
- Backcasting framework
- Future scenarios generator
- Four scenario archetypes
- Deterministic prediction chain
- Hybrid AI forecasting
- Uncertainty mapping toolkit
- Core hybrid prediction engine
- Four-scenario generation
- Backcasting workbench
- Real-time data feed integration (APIs from World Bank, IMF, NASA)
- Collaborative scenario editing (multi-user)
- GPT-5 fine-tuned foresight model
- Mobile companion app (iOS + Android)
- Public leaderboard of driver impact scores
We welcome contributions from futurologists, data scientists, and developers. See CONTRIBUTING.md for guidelines.
"The best way to predict the future is to create it — but first, you need to see all possible futures clearly."