GenPark AI Agent Skill - Temporal Knowledge Graph Event Triplet Store

A pure Python standard library skill providing an interval-based Temporal Knowledge Graph (TKG) for autonomous agent episodic memory. Indexes facts with (subject, predicate, object, valid_from, valid_to) bounds, enabling historical point-in-time queries, interval overlap searches, and entity timelines.

Architecture

graph TD
    A[Incoming Agent Event] --> B[Temporal Triplet Extractor]
    B --> C[Interval Indexer: valid_from, valid_to]
    C --> D[(Temporal Knowledge Store)]
    E[Point-in-Time Query: T_target] --> F{Filter: valid_from <= T <= valid_to}
    D --> F
    F --> G[Historical State Snapshot]

Features

• Temporal Interval Indexing: Rigorous Allen interval logic over discrete or continuous time.
• Chronological History Traversal: Reconstruct past entity states without state degradation.
• Zero Pip Dependencies: Pure Python 3.9+ standard library.

Citations & Ecosystem

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