1. Purpose and scope
This policy explains how Carefully Coded Ltd identifies, prioritises and fixes vulnerabilities, develops software securely and controls changes to live systems. It applies to:
- company devices, accounts and cloud services;
- software we build or maintain for clients, including web platforms, mobile apps, back-end services and databases;
- firmware and software on hardware we supply or install, such as displays and sensor stations;
- third-party libraries and components used in any of the above.
It forms part of our Information Security Policy. Where a client contract sets shorter timescales, the contract takes precedence for that client's systems.
2. Finding vulnerabilities
We identify vulnerabilities through:
- automatic update notices from operating system and software vendors;
- automated dependency scanning of every active code repository;
- security advisories for the languages, frameworks and platforms we use;
- independent penetration tests, as set out in our Penetration Testing Policy;
- reports from clients, users and security researchers (see section 8).
The director checks for new advisories affecting our systems at least weekly.
3. Severity and fix timescales
We rate each vulnerability using the Common Vulnerability Scoring System (CVSS) base score, adjusted for how exposed the system is and what data it holds. The clock starts when we become aware of the vulnerability.
| Severity | CVSS score (guide) | Fix or mitigate within |
|---|---|---|
| Critical | 9.0 to 10.0 | 72 hours |
| High | 7.0 to 8.9 | 14 days |
| Medium | 4.0 to 6.9 | 30 days |
| Low | 0.1 to 3.9 | 90 days, or at the next planned release if sooner |
If a full fix is not available in time, we apply a temporary mitigation (for example disabling a feature, restricting access or adding a firewall rule) within the same timescale and record it. If neither is possible, the director records a risk acceptance with a reason and a review date, and we tell any affected client.
For hardware installed on client sites, where access must be arranged with the client, we apply mitigations remotely where we can and agree the earliest possible on-site date with the client.
4. Patching records
For every Critical or High vulnerability we record the date found, the systems affected, the severity, the action taken, the date fixed and who did it. Medium and Low items are tracked in our issue tracker. The director reviews open items at least monthly, and overdue items are raised at management review.
5. Secure development lifecycle
5.1 Design
- We consider security and privacy at the start of every project, including what data is held, who can access it and how it is protected.
- We record significant security decisions for each project.
5.2 Coding standards
- We design and test against the OWASP (Open Worldwide Application Security Project) Top 10 risks, including injection, broken access control, cryptographic failures and insecure design.
- Input is validated, output is encoded, and database queries are parameterised.
- Data is encrypted in transit using current TLS (Transport Layer Security) and, where it holds personal or sensitive data, at rest.
5.3 Code review
- Every change to a client's production code is reviewed before it is merged. Where possible the reviewer is not the author.
- Where only one person is available, the author completes a written self-review checklist and the change is reviewed by a second person afterwards, within 5 working days.
5.4 Dependencies
- Dependencies come from official package registries and are pinned to known versions.
- Automated dependency scanning runs on every active repository. Findings are fixed within the timescales in section 3.
- Unmaintained dependencies are replaced when found.
5.5 Secrets management
- Passwords, API keys, tokens and certificates are never stored in source code or committed to repositories.
- Secrets are held in a dedicated secrets store or encrypted environment configuration, with access limited to those who need it.
- Repositories are scanned for committed secrets. Any exposed secret is revoked and replaced as a Critical vulnerability.
5.6 Testing and test data
- Changes are tested before release, including security-relevant tests where appropriate.
- Live personal data is not used for testing except as allowed by our Test Data Policy.
- Use of AI coding assistants follows our AI Use Policy.
6. Change control
All changes to production systems follow this process:
- Request: the change is recorded with a description, reason and risk rating.
- Approval: standard low-risk changes, such as routine dependency updates that pass all tests, are pre-approved. Other changes are approved by the director, or by the client where the contract requires it.
- Testing: the change is tested in a non-production environment first.
- Rollback plan: every change has a documented way to undo it, such as redeploying the previous version or restoring a backup.
- Release: the change is deployed, and checked after deployment.
- Record: we keep a record of what changed, when, who approved it and who deployed it. Version control and deployment history form part of this record.
Emergency changes to fix a Critical vulnerability or an incident may be made before formal approval. They must be recorded and reviewed by the director within 2 working days.
Major changes, such as a new system, a new hosting environment or a significant change to authentication, trigger a penetration test under our Penetration Testing Policy.
7. Unsupported systems
Software and hardware that no longer receive security updates are removed, replaced or isolated from the internet and from client data. Where a client relies on an unsupported component, we tell them in writing and agree a plan.
8. Reporting a vulnerability to us
If you think you have found a security vulnerability in our website or in a system we build or run, please email info@carefully.codes with "Security" in the subject line. Include what you found, where, and how to reproduce it.
We will:
- acknowledge your report within 2 working days;
- assess it and tell you our severity rating within 5 working days;
- fix it within the timescales in section 3 and let you know when it is fixed;
- tell the affected client where the vulnerability is in their system.
Please do not access, change or delete data that is not yours, disrupt services, or share details publicly until we have fixed the issue. We will not take legal action against anyone who reports in good faith and follows these guidelines. If a report reveals a possible incident, we handle it under our Incident and Data Breach Response Procedure.