Show HN: CacheCore – semantic agent caching with dependency invalidation

Category: infrastructure

Tags: llm-caching, agent-cost-reduction, semantic-cache

Score: 6.3/10 (Innovation: 6, Technical: 6, Documentation: 4, Utility: 7)

CacheCore is a semantic caching proxy for LLM agents that reduces API costs by caching both exact and semantically similar calls across multi-agent pipelines. Its integration simplicity via OpenAI-compatible endpoint makes it practically valuable for cost-sensitive AI deployments.

Target audience: backend devs

Repository: https://cachecore.it/

View on Hacker News