The engineering questions that matter are rarely about which tool. They are about which failure you are willing to own, and how loudly it will tell you when it happens.
This piece walks through how we think about it at Paycux, what we have changed our minds about, and where the sharp edges are.
Quick refresher on prototype pollution
Quick refresher on prototype pollution is where this gets concrete. Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
What's a gadget chain?
Consider what's a gadget chain?. Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Idempotency is not a nice-to-have in any system with retries. Key on an identifier the caller supplies, store the outcome, and return the same answer to the same key rather than doing the work twice.
- Name the boundary before you cross it
- Measure what users feel, not what is easy to instrument
- Make every retried operation idempotent
- Write down the assumption that would invalidate the design
The Axios chain, step by step
The Axios chain, step by step is where this gets concrete. Idempotency is not a nice-to-have in any system with retries. Key on an identifier the caller supplies, store the outcome, and return the same answer to the same key rather than doing the work twice.
Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Step 1: Pollute the prototype (somewhere else)
Step 1: Pollute the prototype (somewhere else) is where this gets concrete. Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Make the boundary explicit.
Step 2: Axios merges the polluted properties into a request
That brings us to step 2: axios merges the polluted properties into a request. Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Step 3: CRLF injection turns a header into a new request
That brings us to step 3: crlf injection turns a header into a new request. Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Step 4: Steal cloud credentials via SSRF
Step 4: Steal cloud credentials via SSRF is where this gets concrete. Measure before you optimise, then measure the thing users feel rather than the thing that is easy to instrument. A p50 that looks fine while the p99 is unusable is a reporting failure, not a performance one.
Make the boundary explicit. When one part of the system can only talk to another through a named interface, you can change either side without a meeting; when it cannot, every change becomes a negotiation.
Where this leaves us
The pattern repeats across every system we have looked at: the hard part is not the mechanism, it is keeping the mechanism honest as the surrounding assumptions change.
If you are working through the same problem and want to compare notes, the docs cover the mechanics and the console shows the behaviour on your own data.
Everything here, already built
Sign-in, enterprise SSO, directory provisioning, roles and an audit trail behind one API. Start with the quickstart and have a working sign-in this afternoon.