Security by design
Security designed into the whole system
Every EduGap system is designed around layered safeguards, controlled access, clear accountability and careful handling of education information—not security added at the end.
Secure foundation
Core protections applied from the start
The reusable EduGap foundation is intended to give each bespoke system a consistent security baseline. The final controls are then reviewed against the trust's actual information, users and risks.
Encrypted connections
Information is protected while moving between a user's browser, EduGap and approved services.
Protected data storage
Stored information is encrypted and separated using access rules appropriate to the organisation and school.
Managed identity and access
Access is based on authenticated identity, role and scope, with least-privilege defaults.
Protected service credentials
Sensitive credentials remain on protected server services rather than being exposed in the browser.
Security review
Dependencies, permissions and release changes are reviewed for avoidable security weaknesses.
Operational monitoring
Errors and important system events can be monitored so that unexpected behaviour can be investigated.
Defence in depth
No single control carries the whole burden
Protection is applied in layers so that identity, permissions, data handling, application behaviour and day-to-day operation support one another.
- 1Identity: confirm who is requesting access before protected information is returned.
- 2Authorisation: restrict each person to the organisations, schools, records and actions appropriate to their role.
- 3Data: encrypt information, minimise what is held and define suitable retention arrangements.
- 4Application: validate inputs, keep sensitive operations on protected services and maintain audit history.
- 5Operations: review releases, monitor behaviour, manage changes and respond to reported concerns.
Clear responsibilities
What EduGap provides, what is configured, and what is shared
Security works best when responsibilities are explicit. The exact division is recorded for each implementation rather than assumed.
Included foundation
- Encrypted connections and protected storage.
- Authenticated access and least-privilege defaults.
- Separation between ordinary editing and sensitive approval actions.
- Server-side handling of secrets and privileged operations.
- Validation of information submitted to the system.
- Audit history for important record changes.
Configured for your trust
- Roles, school boundaries and approval routes.
- The data fields, reports and integrations that are genuinely required.
- Retention periods and access-review arrangements.
- Which AI-assisted features are enabled and who may use them.
- Release checks for the trust's configured workflows.
Shared responsibility
- The trust controls user access, joiners, movers and leavers.
- Users keep credentials secure and report suspected compromise promptly.
- Both parties agree what information is appropriate to process.
- The trust reviews AI-assisted findings and remains responsible for decisions.
- Incident, continuity and data-protection responsibilities are documented before production use.
Controlled AI
AI assists. People remain accountable.
AI can help explain patterns and prepare work, but it does not receive authority to make the trust's decisions.
- AI outputs are visibly labelled, dated and linked to their available sources.
- AI receives only the information the requesting user is permitted to access.
- A named person reviews an AI-assisted finding before it becomes an agreed action.
- AI cannot independently approve, publish, purchase or communicate externally.
- Enabled AI features are recorded and can be restricted or turned off.
- Uncertainty, missing information and data-quality concerns are shown rather than concealed.
Review before release
A bespoke system should not move into production simply because it functions. Access, data flows, approvals, recovery needs and known limitations are reviewed first.
- 1Define the information, users, roles and decisions the system needs to support.
- 2Review the proposed data flows, access boundaries and supplier dependencies.
- 3Configure and test permissions, validation, approval controls and audit records.
- 4Complete an agreed release review before real organisational information is introduced.
- 5Monitor operation, review access and improve controls as the system and risks change.
Evidence and assurance
Security is maintained, not declared once
The appropriate assurance activity depends on the system, information and risk. EduGap's approach is to agree evidence requirements clearly and avoid claiming assurance that has not been completed.
Operational assurance
Release reviews, access checks, dependency review, monitoring and incident handling form an ongoing cycle. Material changes should trigger a fresh review.
Customer security information
Before production use, EduGap can document the agreed architecture, data flows, responsibilities, subprocessors, retention approach and available assurance evidence.
Security requirements are confirmed for each bespoke implementation. Proposed controls or future assurance work are identified as such until they have been completed and evidenced.
Honest limits
What EduGap does not currently claim
Clear boundaries are part of responsible security communication.
- No current Cyber Essentials, Cyber Essentials Plus, ISO 27001 or other certification is claimed.
- No approval or endorsement by the Department for Education is claimed.
- EduGap does not guarantee legal, regulatory, safeguarding or UK GDPR compliance.
- No third-party connection is described as live unless it has been agreed and verified for that implementation.
- The public demonstration does not contain real schools, people, pupil records or safeguarding cases.
- The system and its AI-assisted outputs do not replace professional judgement or specialist advice.
Common questions
Security questions, answered plainly
Does EduGap need to hold pupil data?
Not automatically. Data is minimised around the agreed purpose. Where pupil-level information is genuinely required, that requirement and its controls must be assessed explicitly.
Can AI make decisions or take action by itself?
No. AI-assisted outputs remain drafts for authorised people to review. Approval, accountability and professional judgement stay with the trust.
Is EduGap currently security certified?
No current certification is claimed. Available evidence and planned assurance activity will be stated accurately during procurement and implementation.
Who can see information across schools?
Only authenticated users whose agreed role and scope permit it. School and trust-wide boundaries are configured for the implementation and should be reviewed regularly.
Have a security or data-protection question?
Ask us directly. We will explain what exists, what would be configured for your trust and what has not yet been independently assured.