Caching Layer Kit

Intelligent multi-tier caching for LLM responses, tool results, and embeddings.

0/4 populated · 0 hits

Configuration

Cache strategy

Cache invalidation

Cache Entries

Similarity search842 hits

L1 (hot)

TTL 3600s

Embedding store320 hits

L2 (warm)

TTL 86400s

Vector index156 hits

L2 (warm)

TTL 86400s

Cold storage45 hits

L3 (cold)

TTL 604800s

Integration Code

import { createCachingLayer } from 'agent-tools-kit/optimization'

const cache = createCachingLayer({ strategy: 'semantic', ttl: 3600 })
agent.use(cache.middleware())