Website security · Raksha guide
Vulnerability Scan vs Penetration Test: What Is the Difference?
A vulnerability scan identifies signals within a defined set of checks. A penetration test investigates whether a weakness can lead to a security failure inside an explicitly authorised scope. The distinction is the method and evidence—not simply the price.
By Raksha · Reviewed 16 November 2026
Compare the four approaches
| Approach | What happens | What it cannot establish |
|---|---|---|
| Raksha passive website scan | Reads public responses for transport, headers, cookies and exposure observations. | No login journey, private database review or proof of every exploit path. |
| Guided live pentest | Supported active probes run against approved targets within a signed engagement. | Not a complete assessment of complex business logic or all attack classes. |
| Autonomous pentest | An agent follows leads using supported probes, with an action trail and request budget. | Not unrestricted exploration or a substitute for a human-led assessment. |
| Human-led engagement | A tester investigates agreed workflows and attack paths; depth depends on the contract. | Still limited by scope, time, access and the methods agreed with the provider. |
These descriptions explain Raksha’s modes, not a claim that every vulnerability scanner is passive. Other scanners can send active requests; check their settings and permission requirements.
The authorised testing boundary
Only test systems you own or have explicit permission to assess. Raksha active testing requires a signed engagement and verified ownership. Agree targets, paths, methods, excluded systems, testing windows and request budgets before a run.
Staff-admin access does not bypass pentest ownership verification. Scope checks and cancellation checkpoints constrain future requests; stopping a test cannot undo requests already sent. Coordinate with the system owner and have a recovery plan.
A signal is not the same as an exploit
A missing security header is a configuration observation. To call a result confirmed, Raksha requires the actor, entry point, failed control, trust boundary, affected resource and observed result. An accessible endpoint or an HTTP 200 response alone is not proof of a vulnerability.
For example, a cross-user access failure requires evidence that one account obtained a resource belonging to another account—not merely that the endpoint exists. A passive public scan cannot answer that question by itself.
Choose based on the question
- Check public configuration after a deployment: start with a passive scan.
- Review database permissions or code: connect the relevant inside-out review.
- Investigate supported active attack classes: create an authorised engagement.
- Assess purchases, approvals, account recovery or complex workflows: scope a human-led assessment.
Use reports and coverage together. No findings means none were observed by the completed checks; skipped, blocked or failed checks are not passes.
See Raksha’s capabilities and current limitsScanning and pentesting work together
Recurring scans can help detect changes between deeper assessments. A pentest provides evidence for a point in time; neither replaces access reviews, secure development, incident response or operational monitoring. Compliance packs map evidence to controls, but are not a certification.
Understand penetration testing cost and deliverablesStart with a site you are authorised to test
A free passive check needs no card. Sign-in, email confirmation, ownership rules and usage limits apply. It is not an unrestricted penetration test.