HomeResources
Proven findings, scanners and manual pentests, compared.
Each approach has a place. Here is an honest look at what each one gives you, and where proof changes the picture.
Three approaches, six questions.
General characteristics of each approach, not a review of any product. Individual tools and testers vary.
| Question | Automated scanners | Manual pentests | Proven findings (Nullgap) |
|---|---|---|---|
| Time to results | Minutes, on every change. | Scheduled ahead, then days to weeks for the report. | A run you start yourself, with the price shown before it runs; PR review on every pull request. |
| False positives | Often many: pattern matches flag code that may not be exploitable. | Usually few, because a person checks each issue. | Only leads that pass four checks are reported; the rest are listed as ruled out. |
| Proof | Typically a rule match and a line number. | Depends on the tester; often screenshots or written steps. | Attack, repeat, harmless control and independent check, with the requests attached so anyone can replay. |
| Fix verification | The rule stops matching on the next scan. | A retest, often booked and paid separately. | The original exploit runs again against the fix on a fresh checkout; a free retest within 60 days. |
| Cost model | Usually per seat or per repository, a subscription. | Usually per engagement, by days of effort. | Per developer for PR review, per pentest by code size, published in the open. |
| Coverage | Broad and consistent for known patterns; weak on logic and access control. | Deep on what the tester has time for; strong on business logic. | Code and running apps under one standard, including access control across up to five accounts and AI apps; leads it can’t prove aren’t reported. |
They work best together.
Broad, fast hygiene
Good for catching known patterns early and cheaply. Expect to spend time triaging what they flag.
Deep, human judgement
Good for novel business logic and scoped engagements. Expect lead time and a point-in-time report.
Only what can be exploited
Good when your team’s time is the bottleneck: fewer, real findings with proof, fixes and fix verification.
Add a human to a proof
Nullgap’s expert review puts a security researcher’s eyes on a finding or a whole report when you want both.
Read next
Questions
Does Nullgap replace a manual pentest?
For many teams it covers what they need between engagements, with proof and fix verification. If you need a person’s judgement, add expert review, or use both.
Does Nullgap find everything a scanner finds?
No, and that’s deliberate. A scanner reports what might be wrong. Nullgap reports only what it can prove, and lists the leads it tested and ruled out.
Can we use the report for audits?
Yes. Pentests produce an attestation letter, an executive summary, a full report and an evidence bundle per finding.
Get your first proof today.
Connect a repository and watch Nullgap find, prove and fix. You only ever see what it could prove.
Free 14-day trial · No card · or book a demo with a security engineer