Measure token savings per AI coding agent, optimize context, and share a live local knowledge graph across 16 CLI clients.
Open on GitHub