GenPark AI Agent Skill - Hierarchical Community Detector

A pure Python standard library skill implementing fast modularity optimization for community detection over large agent memory networks. Partitions heterogeneous knowledge graphs into cohesive topical clusters.

Architecture

graph TD
    A[Agent Memory Graph Edges] --> B[Adjacency & Degree Computation]
    B --> C[Initial Community Assignment: 1 node = 1 comm]
    C --> D[Greedy Modularity Gain Maximizer]
    D --> E{Delta Q > 0?}
    E -->|Yes| F[Reassign Community & Repeat]
    E -->|No| G[Converged Thematic Graph Partitions]

Features

• Greedy Modularity Maximization: Effective partitioning without external heavy graph packages.
• Thematic Subgraph Retrieval: Speeds up RAG by searching within topic communities.
• Zero Pip Dependencies: Standard Library Only.

Citations & Ecosystem

• Platform: GenPark AI
• MCP Registry: GenPark MCP Hub