version: "1.0.0" name: upstash-redis title: Upstash Redis category: Backend & Data description: Use to add serverless Redis (caching, counters, sessions) or rate limiting over HTTP from edge/serverless runtimes. tags: [redis, cache, serverless, edge, rate-limiting, kv] official_docs: https://upstash.com/docs/redis sources:

• https://upstash.com/docs/redis/overall/getstarted
• https://upstash.com/docs/redis/sdks/ts/getstarted
• https://upstash.com/docs/redis/sdks/ratelimit-ts/gettingstarted

last_verified: 2026-08-10

Upstash Redis — Skillship

Serverless, HTTP-based Redis that works from edge/serverless runtimes (Vercel, Cloudflare Workers,

Lambda) where raw TCP is restricted. Pay-per-request, scales to zero.

🧭 When to use this skill

• Use when: you need caching, counters, sessions, queues, or rate limiting in serverless/edge code.
• Use when: your runtime can't open TCP sockets (edge) — Upstash speaks HTTP/REST.
• Don't use for: relational data or transactions across tables (use Postgres).

⚡ Quickstart

1. Install

npm install @upstash/redis
# For rate limiting:
npm install @upstash/ratelimit

2. Configure env (Console → your DB → REST API)

UPSTASH_REDIS_REST_URL=https://<id>.upstash.io
UPSTASH_REDIS_REST_TOKEN=<token>

3. Minimal working example

import { Redis } from "@upstash/redis";
const redis = Redis.fromEnv(); // reads UPSTASH_REDIS_REST_URL/_TOKEN
await redis.set("counter", 0);
await redis.incr("counter");
const value = await redis.get<number>("counter");

🧩 Common recipes

Recipe: Cache with TTL

await redis.set("user:42", JSON.stringify(user), { ex: 60 }); // expires in 60s
const cached = await redis.get<string>("user:42");

Recipe: Data structures

await redis.zadd("scores", { score: 1, member: "team1" });   // sorted set
await redis.lpush("queue", "job-1");                          // list
await redis.hset("people", { name: "joe" });                 // hash
await redis.sadd("animals", "cat");                          // set

Recipe: Rate limiting (sliding window)

import { Ratelimit } from "@upstash/ratelimit";
import { Redis } from "@upstash/redis";
const ratelimit = new Ratelimit({
  redis: Redis.fromEnv(),
  limiter: Ratelimit.slidingWindow(10, "10 s"), // 10 requests / 10s
  analytics: true,
});
const identifier = userId ?? ip;               // per-user or per-IP
const { success } = await ratelimit.limit(identifier);
if (!success) return new Response("Too many requests", { status: 429 });

Recipe: Edge runtimes (Cloudflare Workers / Vercel Edge)

import { Redis } from "@upstash/redis/cloudflare"; // edge-safe import
// Ensure background analytics/replication finish before the runtime exits:
const { success, pending } = await ratelimit.limit("id");
context.waitUntil(pending); // Workers; on Vercel use waitUntil from the Functions API

🚀 Ship to production

• [ ] UPSTASH_REDIS_REST_URL and UPSTASH_REDIS_REST_TOKEN set in the host env (Vercel/Workers secrets).
• [ ] Rate-limit background work awaited via waitUntil(pending) in edge/serverless.
• [ ] Choose a primary region near your writes; add read regions for global reads.
• [ ] Set TTLs on cache keys so memory/cost don't grow unbounded.
• [ ] Namespace keys (e.g. prefix) if sharing a DB across apps to avoid collisions.

🔐 Security & secrets

• The REST token grants full DB access — store as a secret, never ship to the browser bundle.
• For client-side reads, create a read-only token; never expose the read/write token publicly.

🐛 Common errors & fixes

💰 Pricing gotchas

• Billed per request (command). MultiRegion replication and analytics add background commands per limit() call.

📚 Sources

• https://upstash.com/docs/redis/overall/getstarted
• https://upstash.com/docs/redis/sdks/ts/getstarted
• https://upstash.com/docs/redis/sdks/ratelimit-ts/gettingstarted