@ratelock/local

Zero-dependency, in-memory rate limiting. Perfect for single-process applications.

When to Use

• Single-server or single-process deployments
• Development and testing environments
• When you want zero external dependencies
• When sub-millisecond latency matters

Installation

npm install @ratelock/local

Quick Start

Fixed Window

import { fixedWindow } from '@ratelock/local'

const limiter = await fixedWindow({
    limit: 100,
    windowMs: 60_000, // 1 minute
})

const result = await limiter.check('user:123')
// { allowed: true, remaining: 99, reset: 1716000060000 }

Sliding Window

import { slidingWindow } from '@ratelock/local'

const limiter = await slidingWindow({
    limit: 100,
    windowMs: 60_000,
})

Token Bucket

import { tokenBucket } from '@ratelock/local'

const limiter = await tokenBucket({
    capacity: 10, // Max burst size
    refillRate: 1, // Tokens per second
})

Individual Fixed Window

import { individualFixedWindow } from '@ratelock/local'

const limiter = await individualFixedWindow({
    limit: 10,
    windowMs: 60_000,
})

Using the Result

const result = await limiter.check('user:123')

if (!result.allowed) {
    return new Response('Too Many Requests', {
        status: 429,
        headers: {
            'Retry-After': String(Math.ceil((result.reset - Date.now()) / 1000)),
        },
    })
}

Batch Checks

const results = await limiter.checkBatch(['user:1', 'user:2', 'user:3'])

Cleanup

await limiter.destroy() // Clears internal state

Documentation

Full API reference and guides at ratelock.vercel.app.

License

MIT