Redis logger for prai, a TypeScript library for programmable AI interactions. Version 0.4.16 provides middleware to log model interactions (prompts, completions, metadata) to Redis streams. Supports automatic serialization, configurable stream names, and integrates with the Node.js redis client. Differentiates by being purpose-built for prai's history/step workflow, offering low-latency logging with Redis stream capabilities. Released under the pmndrs organization, actively maintained.
npm install prai-redisNo compatibility data collected yet for this library.
Verified import paths — ran on the pinned version, not inferred.
Initializes a prai model and history, creates a Redis client, sets up Redis logger to stream interactions, runs a step, and disconnects.
Ensure Redis URL is correct and server is accessible before calling redisLogger.
Always await client.connect() before passing client to redisLogger.
Install redis@4 or later and use the modern createClient API.
Change to: import { redisLogger } from 'prai-redis'Ensure client.connect() is awaited before redisLogger, and avoid reusing after client.disconnect()
Add password to REDIS_URL: redis://:password@host:port