Your systems will get tested.
The only question is by whom.

Offensive Security Services

IMPETUS OPS

Impetus Ops Logo

Momentum for your defense

Get Quote In 24h

Questions or want to talk first? Email me.


Focused efforts. No fluff. Actionable results.

Threat actors target your technology and your people, relentlessly. Don't wait for a costly breach to validate your security posture.

Limited services  ·  Limited clients  ·  Quality over quantity.

Have your systems tested by ethical hackers - before the bad ones do.


Services

What I offer

Fewer services, more focus, higher quality. Simple as that.

Web Application Penetration Testing  (WAPT)

YOUR APPLICATION
IS LIVE.
SO ARE THE THREATS.

WAPT WEB APPLICATION PENTEST MANUAL-FIRST · LOGIC-AWARE · ZERO-NOISE
In Brief

A service exposed to the internet is under constant attack. Automated tools cannot interpret your unique business context - they stop at pattern matching. This is a manual-first assessment of your application's core revenue paths, workflows, and authentication flows. The assessment models real user journeys, roles, and edge cases to uncover business logic vulnerabilities that automated tools structurally cannot find.

You get a zero-noise deliverable: no false positives, clear exploit narratives, and prioritized remediation guidance for your developers.

Penetration testing enables organizations to uncover vulnerabilities that might otherwise go unnoticed and assess their potential technical and business impact. By identifying and addressing these weaknesses proactively, companies can significantly reduce the likelihood and potential impact of a successful attack.

Because applications and infrastructures change constantly - new features ship, libraries and cloud services evolve, fresh CVEs emerge (especially with AI as facilitator) - penetration testing should be performed on a regular cadence and after any significant change or security event.

Routine assessments reduce the exposure window between releases, validate that previous fixes remain effective, and surface newly introduced risks before attackers can exploit them.

Tools can't understand business logic. My approach puts the auditor at the center, supported by tools, not vice versa. Automation accelerates routine checks, while manual testing targets complex logic and hidden functionalities - where human insight adds the most value.

Deep exploration of application logic via human reasoning helps uncover how legitimate features might be misused or abused and how minor weaknesses could be chained to affect confidentiality, integrity, and availability.

Standards-backed. Experience-driven. The assessment methodology is grounded in trusted standards like OWASP WSTG, PTES, and CWE. KEV and EPSS scores are consulted to define real-world risk accuracy. Backed by real-world engagements and bug bounty research, findings are interpreted with the judgment that frameworks alone cannot provide.

Testing engagements are structured, yet flexible - starting with pre-engagement, reconnaissance (passive/active), vulnerability identification, exploitation, post-exploitation, reporting, debrief and retest.

01 Pre-Engagement 02 Reconnaissance 03 Vulnerability Assessment 04 Exploitation 05 Post-Exploitation 06 Reporting 07 Debrief 08 Retest

WSTG: Web Security Testing Guide
PTES: Penetration Testing Execution Standard
CWE: Common Weakness Enumeration
KEV: Known Exploited Vulnerabilities
EPSS: Exploit Prediction Scoring System

Threat actors continuously search for vulnerabilities, misconfigured roles, exposed endpoints, exploitable input fields, and any other weaknesses that can be leveraged.

  • "Could a malicious user escalate privileges, leak PII, inject arbitrary commands, or exploit a logic flaw?"
  • "Can a user access another tenant's data, bypass a paywall, or manipulate pricing logic?"
  • "Could an attacker take over a customer account without ever knowing the password?"

These are some of the questions penetration testing answers - before attackers do.

TIME SKILL REQUIRED 2000 2005 2010 2015 2020 2025 NONE LOW MED HIGH EXPERT Exploit Kits Manual · dark web kits Crimeware-as-a-Service Botnets, phishing kits RaaS · Infostealers Ransomware affiliates, MaaS AI / LLMs Malware & exploit gen.

The technical barrier to launching effective attacks has been collapsing for two decades — and AI is accelerating it.


The numbers make the case.

  • In 2024, over 40,000 CVEs were published (+38% vs 2023). [1]
  • On average, 108 new vulnerabilities disclosed daily. [2]
  • 768 CVEs were exploited in the wild in 2024 - up 20% year-over-year. [3]
  • 23.6% of Known Exploited Vulnerabilities were exploited before public disclosure. [3]
  • Time from CVE disclosure to confirmed exploitation dropped from 2.3 years (2018) to 20 hours (2026). [4]


An unsecured system could lead to costly consequences:

  • Average global cost of a data breach hit $4.44M in 2025. [5]
  • In the US, average breach costs reached $10.22M in 2025. [5]
  • Healthcare PII breaches average $7.42M per org in 2025. [5]

Takeaway: paradigm change. Consider your exposed surface as constantly under attack.

A Penetration Test can be conducted from different perspectives based on the prior knowledge level of the auditor.

KNOWLEDGE ACCESS LEVEL LOW MEDIUM HIGH LOW MED HIGH BB BLACK BOX External attacker GB RECOMMENDED GREY BOX Authenticated user WB WHITE BOX Full source access
  • Black Box (External): No prior knowledge or credentials. Simulates an external attacker.
    • "What happens if a stranger attacks my app without credentials?"
  • Gray Box (Authenticated - Recommended): Credentials provided. Simulates insider threats or compromised accounts. Deep exploration of business logic.
    • "What could a malicious user do using/abusing exposed functionalities?"
  • White Box (Full Access): Complete access to source code, architecture, and documentation.
    • "Are there vulnerabilities buried in my codebase that no external scan would ever surface?"
  • 1-hour debrief call: every finding explained in plain language for both technical and non-technical stakeholders.
  • One full retest to verify all findings.
  • 30-day feature review: one targeted reassessment of an isolated feature or fix shipped post-delivery.
  • Findings tracker: asset, vulnerability, severity, owner, remediation status - ready to hand directly to your development team.
  • Detailed remediation guide for every finding.
  • Short exploitation videos* for complex vulnerabilities.
  • Attack path diagrams*: high-level visual maps of how individual vulnerabilities chain into a realistic multi-step attack.
  • Post-Exploitation Session*: if initial access is identified during testing and the client agrees, the environment is explored to evaluate the full extent of potential impact.

*Where applicable.

01 24h Quote in 24h 02 call Intro Call scope & align 03 docs Formalities agreement 04 auth Authorization written + access 05 test Testing per agreed scope 06 call Debrief report + call 07 done Retest verify fixes
  • Quote in 24h - You share a short description of your web application (3-minute form). Within 24 hours you receive an initial proposal with scope, timelines, and indicative pricing.
  • Intro Call - A short call to refine the scope, align expectations, and clarify technical details. If the scope changes on the call, the quote is revised accordingly - up or down. You never move forward on a number that doesn't reflect the actual work.
  • Formalities - The service agreement is finalized so everyone knows exactly what is included and what is not.
  • Authorization to test - You provide written authorization and required access (URLs, test accounts, VPN or IP whitelisting). This keeps the engagement fully legal.
  • Testing starts - The web application penetration test begins according to the agreed scope and dates. You receive interim updates if critical issues are found, so your team can start fixing before the final report.
  • Debrief - You receive the final report and walk through every finding on a call. Severity, impact, and remediation priorities are explained clearly for both technical and non-technical stakeholders.
  • Retest - Once your team has addressed the findings, one full retest is included at no extra cost to verify that vulnerabilities have been properly fixed and no regressions were introduced.
Pricing
€2,000*
Delivered in 5–7 business days

Covers one application, black-box, standard complexity. Final pricing is scoped individually based on: test type (black/gray-box), number of user roles and authentication flows, features, critical workflows, integrations, API surface, number of environments.

*VAT or local taxes may apply.

Average cost of a data breach    $4.44M   ·   This engagement    €2,000

No call needed to get a quote

Base pricing already visible - no surprises. Sample report available - judge the work before committing. Tailored quote in 24h - no call required.

If the numbers or the approach don't fit, no time lost on either side.


Deliverables

No surprises

You need to know what you're getting before you get it.

Sample deliverable

See what your final report looks like

Clear executive summary, technical findings, proof of impact, remediation guidance, and more.

  • Executive summary for technical and non-technical stakeholders
  • Step-by-step vulnerability reproduction with evidence
  • Impact explanation and remediation guidance
Red pill or blue pill - choose reality
What You Get
  • Tools in the hands of a reasoning human, not the opposite.
  • No bloated reports - straight to the point.
  • Reproducible proof of concepts.
  • Proper impact and severity evaluation.
  • Specific recommendations.
  • Well-organized findings tracker for easier vulnerability management.
  • Continuous feedback.
  • A human to communicate with.
  • Direct access to your tester - no middlemen.
What You Don't Get
  • Vulnerability assessment scan sold as a pentest.
  • Reports filled with walls of text.
  • Unintelligible proofs of concept without descriptions.
  • Reverse tabnabbing reported as High severity.
  • One-liner remediation.
  • Reports that lack structure for actionable vulnerability management.
  • Ghosting.
  • A chatbot to communicate with.
  • Three people CC'd on every email, none of them your tester.

Credentials

Certifications

Offensive Security is like fighting - you must keep training to stay sharp. Only practical certs. Trusted by the industry. Recognized by vendors.

OSCPOffensive Security Certified Professional (OSCP)
CRTOCertified Red Team Operator (CRTO)
VHLVirtual Hacking Labs Advanced+ (VHL+)
CRTPCertified Red Team Professional (CRTP)
eJPTeLearn Junior Penetration Tester (eJPT)
MCRTAMulti-Cloud Certified Red Team Analyst (MCRTA)
CRT-IDCertified Red Team Infrastructure Developer (CRT-ID)
CRTACertified Red Team Analyst (CRTA)
AD-RTSActive Directory Red Teaming Specialist (AD-RTS)

Real-World Impact

Acknowledgments

Certifications and formal recognition are a starting point, but real-world findings and acknowledgments require a different perspective.

CVE-2021-37596XSS - Telegram Web K Alpha 0.6.1
CVE-2021-40532Chat DoS - Telegram Web K Alpha <0.7.2
CVE-2021-23439XSS - file-upload-with-preview <4.2.0
CVE-2021-23562XSS - plupload <2.3.9
CVE-2021-37504XSS - jQuery-Upload-File v4.0.11
CVE-2021-42244XSS - Notimoo v1.2
Command InjectionCommand Injection - Zentaopms
Session FixationSession Fixation - Zentaopms
Cross-Site ScriptingXSS - Zentaopms
Improper AuthorizationImproper Authorization - Zentaopms
CSRFCross-Site Request Forgery - Zentaopms
Cross-Site ScriptingReflected Cross-Site Scripting - emoncms
Cross-Site ScriptingDOM Cross-Site Scripting - emoncms
UndisclosedSalesforce
UndisclosedKonica Minolta

Feedback

From Clients &
Collaborators

Direct client outcomes, plus feedback from security peers and collaborators I've worked with closely.

“A targeted security assessment delivered findings our development team could act on immediately.”

President, Student Travel Platform (Go2RAIL)


Proof

By the Numbers

With so much marketing noise, choosing who to trust isn't easy.

6

CVEs Discovered

6+

Years in Offensive Security

100+

Applications Tested


FAQ

Frequently Asked
Questions

You get the actual tester, not a project manager. No account managers, no outsourced analysts, no vulnerability assessment dressed up as a pentest. You work directly with me from scoping to final report. I take on a limited number of engagements at a time.

Startups, scale-ups, and companies heading into a funding round, audit, or product launch - or any team that needs an offensive perspective on their security posture.

A standard black-box assessment typically takes around 5-7 business days from kickoff to final report. More complex scopes are estimated individually and defined in your proposal.

Fill in a short 3-minute form. You'll receive an initial proposal within 24 hours. If needed, a short scoping call follows to refine scope and confirm pricing before anything is signed.

Yes. A remediation retest is available after your team has addressed findings. This is already included in the initial quote. Furthermore, if you ship a feature change within 30 days of the assessment, a limited retest scoped to that feature is included.

Automated scanners match patterns - they can't reason. I manually test authentication flows, business logic, and access controls, chaining findings the way a real attacker would to uncover what scanners structurally miss.

No assessment guarantees complete protection. What you get is a clear picture of your current attack surface, prioritized findings, and actionable steps that significantly reduce real-world risk. But risk doesn't stand still - new features, integrations, and updates introduce fresh exposure every cycle. One pentest is a snapshot. Test regularly: quarterly for high-risk apps, or after every major release. The goal is continuously shrinking your attack surface faster than it grows.


About

About Me

Michele Di Stefano

I started in security consultancy, working across client environments with different stacks, industries, and teams. Rarely did two engagements look the same.

Outside client work, I've reported vulnerabilities to Telegram, Flickr, Salesforce, Konica Minolta, Kaseya, and others through bug bounty programs and responsible disclosure - a few of those ended up as CVEs.

For a period I also worked in incident response. When something has already gone wrong, the pressure is real and the stakes are obvious. That experience stays with me every time I write a report or walk a team through findings.

I’ve also mentored junior testers, helping them develop practical testing methodologies, think critically about attack paths, and move beyond checklist-driven assessments.

Currently, my focus is bug bounty research and freelance penetration testing.

Break it. Document it. Help you fix it.


Your systems will get tested The only question is by whom

Get Quote In 24h

Questions or want to talk first? Email me.