README.md

metadata

title: CASCADE-LATTICE Chess
emoji: ♟️
colorFrom: indigo
colorTo: purple
sdk: docker
app_file: app.py
pinned: false
license: mit
tags:
  - cascade-lattice
  - causal-intelligence
  - provenance
  - merkle
  - explainable-ai
  - chess
  - 3d-visualization

♟️ CASCADE-LATTICE Chess

Causal Intelligence Framework — Live Demo via 3D Chess

What is this?

This is a live demonstration of the cascade-lattice package - a causal intelligence framework for tracking causation, detecting anomalies, and providing cryptographic provenance for AI decisions.

Chess is just the vehicle - the real star is cascade-lattice.

🎮 Features

• 3D Chess Board with Three.js visualization
• Adversarial Decision Graph showing White (top) vs Black (bottom) decision flow
• Full Decision Matrix - see ALL candidates the AI evaluated at each move
• Event Inspector - click any decision or alternative to see complete metadata
• Merkle Provenance - cryptographic chain of every event
• Timeline Navigator - scrub through the complete decision history

🔬 Showcased cascade-lattice Features

⛓️ CausationGraph - Every move registered with causal links

• causation_graph.add_event() / add_link()
• get_root_events() / get_leaf_events()
• get_causes() / get_effects() / find_path()

🔍 Tracer - Bidirectional causation tracing

• tracer.trace_backwards() / trace_forwards()
• tracer.find_root_causes()
• tracer.analyze_impact()

📊 MetricsEngine - Quantification and anomaly detection

• metrics_engine.record() / get_metric()
• health_summary() / anomalies()

🔐 Merkle Provenance - Cryptographic chain of every event

Install cascade-lattice

pip install cascade-lattice

Quick Example

from cascade import CausationGraph, Tracer, Event, MetricsEngine

# Create causation graph
graph = CausationGraph()
tracer = Tracer(graph)
metrics = MetricsEngine()

# Register events with full decision data
event = Event(
    event_id="evt_1", 
    component="chess", 
    event_type="move",
    data={
        "move": "e2e4",
        "all_candidates": [...],  # Full decision matrix
        "merkle": "abc123..."
    }
)
graph.add_event(event)

# Trace causation
root_causes = tracer.find_root_causes("evt_1")
impact = tracer.analyze_impact("evt_1")

---

Built with Dash, Three.js, dash-cytoscape, and cascade-lattice