@ai-sdk-tools/cache weekly downloads statistics

@ai-sdk-tools/cache

Universal caching wrapper for AI SDK tools. Cache expensive tool executions with zero configuration - works with regular tools, streaming tools, and artifacts.

Why Cache Tools?

AI agents repeatedly call expensive tools: • Same API calls across conversation turns (weather, translations) • Heavy calculations with identical parameters (financial analysis) • Database queries that don't change (user profiles, company data) • Streaming tools with complex artifact data (charts, metrics)

Caching provides: • 10x faster responses for repeated requests • 80% cost reduction by avoiding duplicate calls • Smooth agent conversations with instant cached results • Complete data preservation - streaming, artifacts, everything

Installation

Quick Start

Basic Usage (LRU Cache)

Result: First call hits the API, subsequent calls return instantly from cache!

Redis/Upstash Cache (Production)

Standard Redis

IORedis

Universal Redis Support

Works with any Redis client that implements: • get(key: string): Promise • set(key: string, value: string): Promise • del(key: string): Promise • setex?(key: string, seconds: number, value: string): Promise (optional)

Supported clients: • ✅ Upstash Redis - @upstash/redis • ✅ Standard Redis - redis • ✅ IORedis - ioredis • ✅ Redis Clusters - Any cluster client • ✅ Custom Redis clients - As long as they implement the interface

Configuration Options

Multi-Tenant Apps (Context-Aware Caching)

For apps with user/team context, just add getContext to the cache config:

Result: Cache keys automatically include teamId and userId - no collisions between users/teams!

Reusable Cache Configuration

For consistent setup across your app, create a configured cache function:

Streaming Tools with Artifacts

Multiple Tools

Custom Callbacks

Cache Statistics

API Reference

createCached(options?)

Creates a cached function with optional Redis client.

Parameters: • options.cache - Redis client (optional, defaults to LRU) • options.keyPrefix - Cache key prefix • options.ttl - Time to live in milliseconds • options.debug - Enable debug logging • options.onHit - Cache hit callback • options.onMiss - Cache miss callback

Returns: A function that wraps tools with caching

cached(tool, options?)

Basic caching function with automatic context detection.

Parameters: • tool - AI SDK tool to cache • options.cacheKey - Function to generate cache key context • options.ttl - Time to live in milliseconds • options.store - Cache store backend • options.debug - Enable debug logging • options.onHit - Cache hit callback • options.onMiss - Cache miss callback

cacheTools(tools, options?)

Cache multiple tools with the same configuration.

Contributing

Contributions are welcome! Please read our contributing guide for details.

License

MIT © AI SDK Tools

Description from the npm listing.