Audit and shrink your Claude Code startup context. Measures what every skill, plugin, agent, and memory file costs in the system prompt, then reversibly disables the dead weight. No proxy, no compression.

C.O.N.T.EX.T is designed to compress complex, multi-domain conversations into machine-optimized "Carry-Packets." These packets achieve a crystallization point of 0.15 entity/token, ensuring that a receiving model can reconstruct the original context with near-perfect fidelity.

SimpleMem: Efficient Lifelong Memory for LLM Agents — Text & Multimodal

vMLX - Use MLX models easily - JANGQ (GGUF for MLX) - Not dependant on mlx_vlm

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