Does our API expose private data? How to check what your API really returns
A safe, own-system method for finding out whether your API hands back more than it should, and to whom.
Key takeaways
- Settle what you can from your own side first: inventory, configuration and permissions cost nothing to check.
- A self-check covers what you thought to look at; an independent test covers what you did not.
- Keep the record of what was checked, by whom, when, and against which version.
On this page
A safe, own-system method for finding out whether your API hands back more than it should, and to whom.
What the question really asks
Most teams arrive at this with a decision to make rather than a topic to study. It is worth being precise about what you are trying to find out, and about what you would do differently depending on the answer.
NoteA check that only ran against staging has told you about staging. Record the environment and the release version you actually tested.
What you can establish yourself
There is usually something you can settle from your own side before anyone tests anything: an inventory, a configuration, a log, a set of permissions. Start with whichever of those you already hold, because it costs nothing and it narrows the question.
- Write down what you already hold: inventory, configuration, permissions.
- Establish what of it you can verify yourself, and against which environment.
- Note what you could not settle, because that is the part worth testing.
Where the limits are
A self-check tells you about the things you thought to look at. The value of an independent test is the things you did not, which is why the two are complementary rather than alternatives. Neither replaces the other.
WarningA compensating control is not a fix. If something only passes because a filter in front of it blocked the request, record it that way and keep the underlying change on the backlog.
What good evidence looks like
Whatever the conclusion, keep the record: what was checked, by whom, when, and against which version. That record is what an auditor, a customer or your next tester will ask for, and reconstructing it afterwards is far more work than keeping it as you go.
Where this leaves you
Treat the above as a starting point rather than the whole answer. If you want does our api expose private data? how to check what your api really returns assessed against your own systems, a scoping conversation is the fastest way to find out which parts apply to you.
Frequently asked questions
Can we do this ourselves instead of commissioning a test?
Partly. Inventory, configuration and permission checks are yours to run and worth running first. Establishing whether a weakness is genuinely exploitable, safely and with evidence, is where an independent tester earns their fee.
How long does this take?
It depends on scope far more than on anything else. A single application is days; an estate with several cloud accounts and internal networks is weeks. A scoping conversation exists to answer this for your case rather than in general.
What do we get at the end of it?
A report you can act on: each finding with the evidence behind it, what it would take to exploit, and what to change. Ask to see a sample report before you buy — every provider structures theirs differently.
Sources
- Web Security Testing GuideOWASP Foundation
- Application Security Verification StandardOWASP Foundation
- SP 800-115, Technical Guide to Information Security Testing and AssessmentNIST
Written by
DERISK Security Team
Offensive Security
The testers who run our penetration testing engagements, writing up what they find worth explaining.

Planning a penetration test?
Tell us what you need tested and we'll help you scope it.
Scope a test