Omen (Chinese: 爻) is an open-source strategic reasoning engine powered by Explainable AI (XAI). It combines ontological modeling of strategic phenomena with counterfactual analysis of uncertainty, delivering verifiable, traceable, and explainable insights for decision-makers.

Concepts | Quick Start | Case Templates | Roadmap

🪄 Capabilities

Analyze, Simulate, Explain.

Omen does not predict the future. It is a reasoning engine built for complexity. By mapping causal chains and logical dependencies, it generates replayable, comparable branching paths, revealing weak signals, critical points, and evolving ecosystems, and helping decision-makers gain clarity in complexity:

• 🔄 Substitution Logic: Which technology will replace another under what critical conditions?
• 🛡️ Capability Evolution: Which core capabilities will be enhanced first, and which will coexist long-term?
• 🏆 Strategy Wins: Which strategy combinations are more likely to win the market, capital, and developer ecosystem?
• ⏳ Time Windows: When is the optimal timing for in-house development, alliances, M&A, or contraction?

Through explainable reasoning chains, Omen reveals how technological evolution reshapes markets, helping strategic decisions decode the omens from the chaos.

✨ Core Features

📊 Typical Outputs

A complete reasoning session typically answers the following questions:

• Substitution? Will the new technology completely replace the old one, or form a complement?
• Time Window? When is the specific time window for substitution or turning points?
• Key Drivers? Which variables (e.g., cost reduction speed, API compatibility) are the decisive factors?
• Winners and Losers? Which entities suffer first, and which benefit unexpectedly?
• Strategy Effectiveness? Under what circumstances is an "open ecosystem" superior to "vertical integration"?
• Endgame Form? Does it move towards monopoly, oligarchic balance, or fragmented coexistence?

🚀 Quick Start

🏗️ Installation

Environment requirements: Python 3.12+ with pip package manager.

git clone https://github.com/StrategyLogic/omen.git
cd omen
pip install --upgrade pip setuptools wheel
pip install -e .

🌰 View Example

If you want to quickly see Omen in action, a visualized sample case and its results are available in the sample directory. Run:

streamlit run sample/app/scenario_planning.py

Then open http://localhost:8501 in your browser to explore the full strategic reasoning flow.

🎵 Run Built-in Case

If you want to run a complete end-to-end workflow, we have prepared a built-in case simulating SAP's acquisition of Reltio in March 2026. The case document is cases/sap_reltio_acquisition.md, and it can be run end-to-end with the following commands:

Step 1. Analyze

Omen's Analyze module combines strategy methodology and the data pipeline, allowing you to generate strategic insights and machine input artifacts from the source document with a single command.

Situation Analysis

# analyze the built-in case and pack it as "sap" alias
omen analyze situation --doc sap_reltio_acquisition --pack-id sap

This step generates the Situation Artifact and creates a package named sap for consistent use in subsequent steps.

Scenario Planning

Omen v0.1.9 provides deterministic A/B/C scenario planning capabilities.

• Scenario A: Offensive branch
• Scenario B: Defensive branch
• Scenario C: Confrontational branch

You can directly use the sap alias to locate the generated situation artifact from the previous step:

omen scenario --situation sap

This step generates the scenario pack artifact under data/scenarios/sap/ for simulation.

Step 2. Simulate

Omen's simulation engine can reason across different scenarios. Use the scenario pack generated in the previous step to run simulation:

omen simulate --scenario data/scenarios/sap/scenario_pack.json

This step generates reasoning traces and writes the deterministic result to output/sap/result.json.

Step 3. Explain

Omen's explanation module interprets simulation outcomes and traces back key decision points and risk items (known unknowns) from the situation artifact to generate decision-ready insights and recommendations:

omen explain --pack-id sap

This step generates a structured explanation artifact at output/sap/explanation.json.

Launch UI

Omen also provides a Streamlit-based UI application for visualizing the full strategic reasoning flow.

streamlit run app/scenario_planning.py

End-to-End Flow

More details

You can click on each panel on the page to inspect the full chain of outputs from source document to situation artifact, scenario artifact, simulation result, and explanation artifact.

👥 Target Audience

Omen is built for the following roles:

• Technology Strategy Teams
• Product & Platform Leads
• AI Infrastructure Researchers
• Open Source Ecosystem Observers
• Investors & Industry Analysts

🎬 Showcase

Strategic Actor Analyze

• 👤 Elon Musk
• 👤 Jeff Bezos
• 👤 Steve Jobs
• 👤 Jack Ma
• 👤 Chen Jiaxing (me)

Strategic Reasoning Cases

• 🧩 Acquisition: SAP vs Reltio
• 🗺️ Ontology Games: Database vs AI Memory
• ⚔️ Vector Database vs AI Memory

More scenarios are under development (contributions welcome):

• Agent Infrastructure vs Workflow Platforms
• Vertical AI vs General AI Stack
• Open Source Models vs Closed Commercial APIs
• Data Governance vs AI-Native Knowledge Systems

📃 License

Omen is under AGPL-3.0-or-later, the project is developed and maintained by StrategyLogic®.

Note: If you wish to use Omen in a closed-source environment or provide it as a SaaS service without open-sourcing your code, please contact us for a commercial license.

🌟 Vision

Omen aims to become an open strategic reasoning workstation:

It does not output a single answer, but helps people systematically understand how the future branches; and understand which conditions shape the outcome; and understand which actions can change the path.

If you are interested in technological evolution, market substitution, strategic modeling, or multi-agent reasoning, you are welcome to join us in interpreting the omens of this chaotic world together.

Simulate the Signs. Reveal the Chaos.