Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.marshell.dev/llms.txt

Use this file to discover all available pages before exploring further.

Every completed scan gets a risk score from 1 to 99 — never 0, never 100. A perfectly clean target asymptotes to 99; an unreachable one floors at 1 so the gauge always renders.

Bands

BandRange
excellent90–99
good75–89
fair50–74
poor25–49
critical1–24

Components

The total is a weighted sum. Each component has its own 0..100 score and a list of hits — click a row in Score breakdown to expand.
ComponentWeightMeasures
Vulnerabilities40%All vuln-scanner findings + CVE matches, weighted by severity.
TLS15%Certificate validity, cipher strength, protocol version, days to expiry.
Headers15%CSP, HSTS, X-Content-Type-Options, X-Frame-Options, Referrer-Policy, Permissions-Policy.
Exposure15%.git/, .env, phpinfo, accessible backups, secrets in response bodies.
Posture15%WAF, bot manager, DDoS mitigation, CDN.
One critical finding can drop the total by 20+. Perfect headers and TLS will not rescue a target with active SQL injection.

Confidence

Shown next to the gauge:
ConfidenceWhen
high≥80% of tools completed cleanly.
medium50–79%.
lowUnder 50%. Score is unreliable; deltas are not rendered.
low confidence usually means the target’s WAF is blocking the scanner. See Firewall bypass.

Deltas

If the same domain has a prior scan with at least medium confidence, Score breakdown shows arrows per component (↑ +7, ↓ −12). Hover for the reason — e.g. “certificate expires in 14 days, was 21” or “CSP added.”