Method and retest policy
What we do during a test, what we will not do, how we rate what we find and what the included retest covers. This is the text we work to; your contract refers to it.
Standards we follow
We use public standards so that you, your customers and your auditors can check our coverage against something other than our word.
- Web applications: OWASP Web Security Testing Guide, with the OWASP Application Security Verification Standard as the checklist for coverage.
- APIs: OWASP API Security Top 10.
- Mobile apps: OWASP Mobile Application Security Testing Guide.
- Scoring: CVSS version 3.1. Each finding carries its full vector, so you can recalculate the score yourself.
Testing is manual. Scanners and proxies support the work, and every finding in the report has been confirmed by hand.
How findings are rated
Severity comes from the CVSS base score. Where the business context makes a finding more or less serious than its score suggests, the report says so and explains why.
| Severity | CVSS 3.1 | What it usually means |
|---|---|---|
| Critical | 9.0 to 10.0 | An outside attacker can take over the application or its data without help from a user. Reported to you as soon as it is confirmed. |
| High | 7.0 to 8.9 | Serious exposure of data or functions, usually needing an account or one extra condition. |
| Medium | 4.0 to 6.9 | Limited exposure, or a serious one that is hard to reach. |
| Low | 0.1 to 3.9 | Little direct impact, but worth fixing as part of normal work. |
| Informational | Not scored | An observation or a hardening suggestion, not a vulnerability. |
Rules we test under
Authorisation comes first
We test only systems named in a written authorisation signed by someone entitled to approve it. If part of your application runs on infrastructure you do not control, such as a cloud platform or a third-party service, we follow that provider's testing policy and leave out anything it does not allow.
What we do not do
- No denial-of-service testing and no load testing.
- No social engineering or phishing of your staff, unless you ask for it and it is written into the scope.
- No testing outside the agreed time window or against targets that are not in scope.
- No scanning or probing of your systems before the authorisation is signed. That includes the period while we are preparing a quote.
When we stop and call you
- We confirm a critical finding.
- We see signs that someone else has already broken in.
- A test step affects the availability of a production system.
- We reach personal data or other sensitive records. We prove access with the minimum needed and go no further.
What the report contains
- A management summary of one to two pages, written for people who will not read the technical part.
- Scope, dates, accounts and roles used, and anything that was left out and why.
- Each finding with a description, evidence you can reproduce, impact, CVSS vector and score, and a specific fix.
- A table of all findings by severity and status.
- After the retest: the result against each finding, and a letter of attestation.
The letter of attestation states who tested what, when, and to which standard, with a count of findings by severity and their status after retest. It contains no technical detail, so you can share it with customers and auditors.
Retest policy
One retest is included in every fixed scope.
- You can ask for the retest at any time within 30 days of receiving the report.
- It covers the findings in the original report. We check each fix and try the obvious ways around it.
- Each finding is then marked as fixed, partly fixed or still open, with new evidence.
- The report is reissued with the retest results and the letter of attestation is updated.
- New functionality and findings outside the original scope are not part of the retest. Further retests, or a retest after the 30 days, are quoted by time.
How we handle your data
- Test notes, evidence and reports are stored encrypted.
- Reports and credentials are exchanged over a channel agreed on the scoping call, never by plain email.
- Test accounts and credentials you give us are used only for the engagement. We ask you to disable them when it ends.
- Evidence and working files are deleted 90 days after the final report, unless your contract sets a different period. The report itself is kept only if you ask us to keep it.
- We do not name clients publicly. A case study is published only with written permission and after you have approved the text.
Questions about any of this: [email protected]