Skip to content
ZekaForge
Security

From scan to remediation: a worked example

Follow one authorized finding from discovery through a prioritized, verified fix.

A finding is only useful if it ends in a fix you can verify. Here is the full loop on one realistic, defensive example.

1. Authorized scan

After verifying ownership of app.example.com with a DNS TXT record, you run a scan with a safe profile. It completes within scope and rate limits, and every step is written to the audit log.

2. The finding

The report flags a missing HTTP Strict-Transport-Security (HSTS) header and a weak Content-Security-Policy. Severity is medium: not an immediate breach, but it weakens defense in depth against downgrade and injection attacks.

3. AI explanation and priority

The assistant explains, in plain language, that without HSTS a user's first request can be downgraded to HTTP, and that a permissive CSP lets injected scripts run. It ranks the HSTS fix first because it is low-effort and high-value, and drafts the exact header values.

4. The fix

You add the headers at the edge or in the app:

Strict-Transport-Security: max-age=31536000; includeSubDomains; preload
Content-Security-Policy: default-src 'self'; object-src 'none'; frame-ancestors 'self'

5. Verify by re-scanning

You run the scan again. The HSTS finding clears; the CSP tightens from "weak" to "present." The report now shows the improvement, and the audit log records both scans.

6. Why the loop matters

Discovery without verification is guesswork. Closing the loop — scan, explain, fix, re-scan — turns a list of problems into measurable, defensible progress.

Partager

Articles liés

Commencez avec ZekaForge

Créez un compte gratuit et explorez Device Cloud et Cyber AI en quelques minutes.