- Native OpenClaw Integration: Plugs directly into prompt assembly via ClawHub, establishing a Three-Layer Memory architecture for your agent (Context Tree: Deep, structured Knowledge Base; Workspace Memory: Core rules, preferences, and guidelines; Daily Memory: Session Notes)

- Git-like Stateful Memory: Organizes memory into a semantic hierarchy of human-readable, diffable, and git-friendly Markdown files.

- Top Market Accuracy: Achieves an industry-leading 92.2% retrieval accuracy (on both LoCoMo and LongMemEval benchmark), maintain 90% accuracy even with cheap & lightweight model

- Setup: Local-by-default, fully portable for multi-agent teams, and installs via a single safe `curl` script.