Penetration Testing Services

Professional security testing combining autonomous AI scanning with expert manual penetration testing. Based in Sweden, serving clients globally.

Our Services

From automated continuous monitoring to deep manual penetration testing — we offer the full spectrum of external security assessment.

Continuous Security Monitoring

From 249 SEK/month

Autonomous 24/7 monitoring of your external attack surface. Our platform scans daily with 30+ professional security tools, detects new vulnerabilities, and delivers AI-analyzed reports.

  • Daily vulnerability scanning (OWASP Top 10, CVEs)
  • Attack surface discovery (subdomains, ports, technologies)
  • Self-learning Verification Engine (eliminates false positives)
  • AI-powered reports delivered to your inbox
  • Compliance mapping (NIS2, ISO 27001, OWASP)
  • REST API + MCP for AI agent integration
Start Free Trial
PREMIUM

Artisanal Penetration Testing

Project-based pricing

Deep, manual penetration testing by experienced Swedish security professionals, augmented by AI analysis. Goes beyond automated scanning — tests business logic, authentication flows, complex attack chains, and human factors.

  • Claude Mythos AI-augmented reasoning for complex attack chains
  • OWASP Top 10 + business logic testing
  • Authentication & authorization testing
  • API security assessment
  • Executive report with remediation roadmap
  • Retest after remediation included
  • Physical security assessment (on-site, optional)
  • Social engineering / HUMINT testing (optional)
Request a Quote

Web Application Testing

From 15,000 SEK

Focused security assessment of web applications. Covers injection vulnerabilities, cross-site scripting, broken authentication, insecure direct object references, and more.

  • SQL injection, XSS, CSRF, SSRF testing
  • Session management & cookie security
  • File upload & input validation testing
  • Detailed vulnerability report with PoC
Get Started

AI-Built App Security Audit

From 5,000 SEK

Security audit specifically designed for applications built with AI tools (Lovable, Bolt, v0, Cursor). These apps have unique vulnerability patterns that traditional scanners miss.

  • Exposed API keys & environment variables
  • Missing authentication on AI-generated endpoints
  • Default credentials & debug mode detection
  • LLM prompt injection testing
Start Audit

Why Pentesting.se?

30+
Security tools in every scan
4-Layer
Verification Engine (self-learning FP elimination)
1983
Adminor AB — Swedish IT security since 1983

Swedish-Hosted, GDPR Compliant

All data stored on servers in Sweden. Full GDPR compliance. No US Cloud Act exposure.

NIS2 & ISO 27001 Ready

Compliance mapping built in. Map your scan findings directly to NIS2 Directive and ISO 27001 Annex A controls.

AI-Powered Analysis

Every scan passes through our Analysis Engine — 4-pass AI pipeline for triage, correlation, deep analysis, and executive reporting.

AI Agent Integration (MCP)

Connect Claude Code, Cursor, or any MCP-compatible AI assistant directly to your security dashboard.

How It Works

1

Scope & Planning

We define the scope, rules of engagement, and testing schedule. For automated monitoring, just add your domain. For manual pentesting, we plan the engagement together.

2

Reconnaissance & Discovery

Our autonomous agents discover your full external attack surface — subdomains, IPs, ports, technologies, cloud resources. Context Engine maps relationships between assets.

3

Vulnerability Assessment

30+ security tools scan for vulnerabilities. The Verification Engine eliminates false positives through 4 layers of detection — only verified findings reach your report.

4

Manual Testing (Premium)

For artisanal pentests, our senior security consultants conduct deep testing — authentication bypass, business logic flaws, API abuse, injection chains — augmented by AI tooling and verified through manual exploitation.

5

Reporting & Remediation

Detailed findings with CVSS scores, evidence, and step-by-step remediation. Executive summary for management. Retest after fixes to confirm resolution.

Frontier AI

Claude Mythos — on our radar

Anthropic's most capable model found thousands of zero-days across every major OS and browser. We are following Mythos closely and will evaluate integration as access broadens — while already building on the most capable AI models currently available.

Learn more →

Secure Your Attack Surface

Start with a free security health check. No account required.

Adminor AB — Swedish IT infrastructure & security since 1983