MeshFlow: Topological Sudoku Matrix

Stop Backtracking, Start Collapsing.

Welcome to the MeshFlow Sudoku Engine. This is not your traditional Depth-First Search (DFS) algorithm. Instead, it is a living, reactive system where 81 individual nodes (cells) continuously negotiate, entangle, and collapse their probabilistic states until a global consensus is reached.

🌀 Core Mechanics

• Constraint Collapse: Values are not explicitly "calculated." As nodes emit forbidden signals to their neighbors, the probability cloud of each cell shrinks. When only one possibility remains, the wave function collapses into a physical value.
• Topological Forensics: Algorithms fail, but they rarely tell you why. When the engine hits a logical deadlock, it triggers a visual forensic scene:
  • 🚨 Victim Nodes (Red Pulse): Singularities where logical space has been crushed to zero.
  • 🕵️ Suspect Nodes (Orange Glow): Neighbors that aggressively occupied the victim's topological space.
• Limited Self-Healing: A centralized "Judge" node monitors the matrix. Upon detecting a deadlock, the Judge intervenes by forcing suspect nodes to retract and logging their failures (Banned List).

  Note: As the engine does not store a global snapshot for full backtracking, this self-healing is probabilistic and limited. When contradictions are deep-seated within an irreversible path, the Judge will declare a terminal deadlock.

🕹️ Interaction Guide

1. Observe Evolution: Watch as candidate numbers dynamically shrink and values crystallize across the grid.
2. ⚡ Re-Collapse: Wipe all deduced values while preserving the initial puzzle. Witness the system's Strong Eventual Consistency—regardless of the random path taken, it always converges back to the same lowest-entropy state.
3. ↻ Zero-State Deduction: Obliterate the current universe and return to a blank slate. Observe how the engine builds order from absolute chaos through local constraints alone.