Skip to main content
getrequest sits in front of any backend as an API reliability layer — so traffic spikes, downtime, or failed deployments never result in lost requests.

​
The problem it solves

When your backend goes down, requests vanish. There’s no record of what came in, no way to recover what was lost, and your users have already moved on. getrequest fixes this at the infrastructure level — before the request ever reaches your code.

Relay

Every request is intercepted at the edge and forwarded to your backend transparently.

Retain

Every request and response is logged — headers, body, status, timing. Nothing is dropped.

Recover

Replay any captured request with one click after a failed deploy or outage.

Share

Generate a public link to any request/response pair and drop it in Slack or a Jira ticket.
That’s the pitch. Here’s how it actually works.

​
How it works

You create an endpoint in getrequest — no code, no deploy. getrequest gives you a URL with a unique, auto-generated ID:
Point your callers at that URL, and getrequest starts capturing every request that arrives — before it decides what to do with it. What it does next is up to you.

​
Three endpoint types

You can switch between them at any time — zero downtime, no redeployment. Not sure which one you need? Sync vs. Async endpoints walks through exactly when to reach for each.

​
Built-in authentication

Whichever type you pick, you can lock down who’s allowed to use it. Every endpoint can require a credential on incoming requests, and can attach a credential to its own outbound destination call — two independent, per-endpoint settings, nothing to code. See Authenticate endpoints and destinations. That’s the whole model: one URL, a type that decides what happens when a request hits it, and logging that never turns off. The rest of these docs go deeper on each piece — but you’ll understand it fastest by just doing it.

Start in 2 minutes

Create your first endpoint and send a request.