ratelimit-next

Rate limiting for Next.js. Zero dependencies.

Declarative rules, type-safe API, pluggable stores. Works in Edge Runtime, Node.js, and serverless.

Install

npm install ratelimit-next

Quick Start

1. Define rules

// lib/rate-limit.ts
import { createFloodgate } from "ratelimit-next";

export const gate = createFloodgate({
  rules: {
    api: { limit: 100, window: "1m" },
    auth: { limit: 5, window: "15m", algorithm: "fixed-window" },
    uploads: { limit: 10, window: "1h", algorithm: "token-bucket" },
  },
});

2. Protect routes

// app/api/data/route.ts
import { withRateLimit } from "ratelimit-next/next";
import { gate } from "@/lib/rate-limit";

export const GET = withRateLimit(gate, "api", async (request) => {
  return Response.json({ data: "hello" });
});

3. Done

Clients get standard rate limit headers (X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset) on every response and a 429 with Retry-After when limited.

Algorithms

Fixed Window

Divides time into fixed intervals and counts requests per interval. Simple and predictable, but allows bursts at window boundaries.

{ limit: 100, window: "1m", algorithm: "fixed-window" }

Sliding Window (default)

Approximates a true sliding window by weighting the previous interval's count. Smoother than fixed window with minimal overhead.

{ limit: 100, window: "1m", algorithm: "sliding-window" }

Token Bucket

Tokens refill at a steady rate. Allows short bursts while maintaining an average rate. Best for APIs where occasional spikes are acceptable.

{ limit: 100, window: "1m", algorithm: "token-bucket" }

Window Format

Duration strings: "10s" (seconds), "5m" (minutes), "1h" (hours), "1d" (days).

Next.js Middleware

Apply rate limiting globally via Next.js middleware:

// middleware.ts
import { createFloodgate } from "ratelimit-next";
import { createRateLimitMiddleware } from "ratelimit-next/next";

const gate = createFloodgate({
  rules: { api: { limit: 100, window: "1m" } },
});

export default createRateLimitMiddleware(gate, {
  rule: "api",
  paths: ["/api/"],
});

export const config = { matcher: "/api/:path*" };

API Route Protection

Three styles to choose from:

withRateLimit (HOF)

import { withRateLimit } from "ratelimit-next/next";

export const GET = withRateLimit(gate, "api", async (request) => {
  return Response.json({ ok: true });
});

rateLimit (guard)

import { rateLimit } from "ratelimit-next/next";

export async function GET(request: Request) {
  const limited = await rateLimit(gate, "api", request);
  if (limited) return limited;

  return Response.json({ ok: true });
}

gate.limit (throws)

import { gate } from "@/lib/rate-limit";
import { RateLimitError } from "ratelimit-next";

export async function GET(request: Request) {
  try {
    await gate.limit("api", ip(request));
    return Response.json({ ok: true });
  } catch (e) {
    if (e instanceof RateLimitError) {
      return Response.json({ error: "Too many requests" }, { status: 429 });
    }
    throw e;
  }
}

React

Client-side components for rate-limit-aware UIs:

import { RateLimited, useRateLimit, RateLimitProvider } from "ratelimit-next/react";

// Wrap your app (optional — configures the check endpoint)
<RateLimitProvider endpoint="/api/rate-limit">
  <App />
</RateLimitProvider>

// Declarative
<RateLimited rule="api-calls" fallback={<p>Too many requests. Please wait.</p>}>
  <MyForm />
</RateLimited>

// Hook
function SubmitButton() {
  const { allowed, remaining, check } = useRateLimit("api-calls");
  return (
    <button disabled={!allowed} onClick={() => { /* submit */ await check(); }}>
      Submit ({remaining} left)
    </button>
  );
}

The React components call GET /api/rate-limit?rule=<name> to check status. Implement this endpoint with your floodgate instance:

// app/api/rate-limit/route.ts
export async function GET(request: Request) {
  const url = new URL(request.url);
  const rule = url.searchParams.get("rule") ?? "api";
  const result = await gate.check(rule);
  return Response.json(result);
}

Stores

Memory (default)

Zero-config, works everywhere. State is lost on restart. Good for development and single-instance deployments.

import { createFloodgate, MemoryStore } from "ratelimit-next";

const gate = createFloodgate({
  rules: { api: { limit: 100, window: "1m" } },
  store: new MemoryStore(), // this is the default
});

Redis

For multi-instance deployments. Requires ioredis.

import Redis from "ioredis";
import { createFloodgate } from "ratelimit-next";
import { createRedisStore } from "ratelimit-next/adapters/redis";

const gate = createFloodgate({
  rules: { api: { limit: 100, window: "1m" } },
  store: createRedisStore(new Redis(process.env.REDIS_URL)),
});

Vercel KV

For Vercel deployments. Requires @vercel/kv.

import { kv } from "@vercel/kv";
import { createFloodgate } from "ratelimit-next";
import { createVercelKVStore } from "ratelimit-next/adapters/vercel-kv";

const gate = createFloodgate({
  rules: { api: { limit: 100, window: "1m" } },
  store: createVercelKVStore(kv),
});

Custom Key Resolvers

By default, rate limits are keyed by IP address. Customize this:

const gate = createFloodgate({
  rules: { api: { limit: 100, window: "1m" } },
  keyResolver: (request) => {
    // Key by API token
    return request.headers.get("Authorization") ?? "anonymous";
  },
});

Or pass keys directly:

const result = await gate.check("api", userId);

API Reference

createFloodgate(config)

Returns a Floodgate instance with:

Method	Description
check(rule, key?)	Check limit, returns RateLimitResult
limit(rule, key?)	Check limit, throws RateLimitError if exceeded
reset(rule, key)	Clear limit for a key
headers(result)	Generate HTTP headers from a result

RateLimitResult

{
  allowed: boolean;
  limit: number;
  remaining: number;
  reset: number;       // timestamp ms
  retryAfter?: number; // seconds (present when blocked)
}

License

MIT

Part of sathergate-toolkit

This package is part of the sathergate-toolkit — an agent-native infrastructure toolkit for Next.js. All packages work independently or together.

See Also

• shutterbox — Image processing pipeline (npm i shutterbox)
• flagpost — Feature flags with percentage rollouts (npm i flagpost)
• notifykit — Unified notifications via Twilio, Resend, SNS (npm i notifykit)
• croncall — Serverless-native cron job scheduling (npm i croncall)
• vaultbox — AES-256-GCM encrypted secrets management (npm i vaultbox)
• searchcraft — Full-text search with BM25 scoring (npm i searchcraft)