Eight steps

Process

From first reply to invoice: eight steps, in this order, and no testing before the third.

  1. Step 01

    Send

    You
    You send your site and your email.
    Us
    We reply within 2 business days.
  2. Step 02

    15-minute scoping call

    You
    You present the scope and your constraints.
    Us
    We ask the technical questions and may decline the mandate.
  3. Step 03 / Gate: closed until signed

    Authorisation

    You
    An empowered person signs the written authorisation and you prove control of the assets (DNS TXT record or similar). If a third-party host is involved, you obtain its consent.
    Us
    We provide the NDA, the contract and the authorisation template. No testing before this step.
  4. Step 04

    Frozen grid and mission cap (30 days, adjustable)

    You
    You approve the grid and the cap.
    Us
    We write them into the contract.
  5. Step 05

    Test window

    You
    You agree the dates with us and flag anything sensitive.
    Us
    We declare our source IP addresses and follow the rules of engagement.
  6. Step 06

    Reports

    You
    You receive each flaw as it is confirmed.
    Us
    Each report contains the CVSS vector, reproduction steps, impact and fix.
  7. Step 07

    Your triage

    You
    You triage within 5 business days.
    Us
    Severity disagreements are settled on the CVSS vector. With no answer by the deadline, the flaw is accepted, as the contract provides.
  8. Step 08

    Invoice

    You
    You pay the validated flaws within 30 days.
    Us
    We invoice only validated flaws. Zero flaws: zero invoice, and a coverage report.

Rules of engagement

What if the test breaks my production?

A test can always have a side effect. Here is what lowers the risk, and what happens when a system reacts badly. All of it is written in the contract.

  1. No denial of service.

    No load testing, no saturation, no attack whose purpose is to bring a service down. If a test could have that effect, we describe it in the report without running it.

  2. Minimal proof.

    We show that the flaw exists: one record read, one screenshot. We do not extract your data, and we prefer to work with the test accounts you provide.

  3. Sensitive data: we stop.

    If we come across sensitive data, we end that test and tell you.

  4. You recognise our traffic.

    Our source IP addresses are declared before the test window starts. Your team knows where we come from and can cut us off at any time.

  5. You choose the moment.

    Dates are agreed together, and you tell us what is sensitive. We agree with you whether the test targets production, staging or both.

  6. Only what is written.

    Only the assets listed in the authorisation are tested. Nothing else, even if it looks connected.

  7. If it reacts badly, we stop.

    Abnormal behaviour from your system, a service slowing down: we suspend the test, tell you, and resume only when you decide.

What you receive

For each flaw and at the end of the engagement.

  1. One report per flaw: CVSS 3.1 vector and score, reproduction steps, impact, recommended fix

  2. An invoice limited to validated flaws

  3. If there are no flaws: a coverage report stating what was tested

Fine by you? Leave us your site and email.

Asking to be contacted takes a few seconds and commits you to nothing.