PrismLib – semantic LLM cache and cluster mesh that cuts token spend

**Brief Summary** PrismLib is an open‑source semantic cache and cluster‑mesh layer for large language models that reduces token consumption by re‑using previous