Permission-aware destinations
Navigation and dashboard destinations can be filtered through effective page permissions.
Trust & security
Control access, preserve scope and keep important decisions traceable.
OptiToolkit applies permission-aware navigation, organisation and data scope, secure-session and request safeguards, output escaping, audit capabilities and explicit AI verification principles across the platform.
Restrictive defaultsTenant-aware data scopeAuditabilityAI verification
Access and data scope
Show each user the pages and records their role permits.
Access controls are part of the server-side decision about what a user can read, change, export, analyse or pass into AI-assisted workflows.
Organisation boundaries
Organisation-owned records retain explicit organisation scope as a defence-in-depth control.
Department and relationship scope
User and department limits can be applied using the relevant entity relationships.
Restrictive defaults
Missing or invalid context should deny access or return no data.
Connected workflow
Protect the session, request, data and output boundaries.
Use shared server-side helpers and restrictive handling across state-changing and data-sensitive workflows.
- 1
Authenticate
Start and validate the secure user session.
- 2
Authorise
Evaluate page permissions and effective data scope.
- 3
Validate
Check method, content type, identifiers, lengths and expected values.
- 4
Operate
Use prepared queries and scoped services for reads and writes.
- 5
Escape
Encode database, user, file, API and AI-derived output.
- 6
Audit
Record stable metadata and support effective-access diagnostics.
Application safeguards
Protect requests and outputs without weakening usability.
The platform uses shared helpers for secure sessions, CSRF protection, output escaping, API request validation and database access patterns.
- Secure session handling for authenticated workflows
- CSRF checks on state-changing forms
- Expected method and content-type validation
- Prepared statements and parameter binding
- HTML escaping for untrusted output
- Defensive upload, external-call and error handling
Platform capabilities
Support review of activity, access and user experience.
Use operational controls across administration, audit, AI and accessible interaction.
Audit and event review
Review available audit, login and operational metadata where configured.
Security diagnostics
Explain effective access decisions and role configuration.
Export controls
Apply capability checks before sensitive operational exports.
Accessible interaction
Use semantic structure, keyboard controls, visible focus and reduced motion.
Scoped analytics
Apply permission and organisation scope to analytics and AI data access.
No unsupported badges
Avoid certification and compliance claims without approved evidence.
AI safeguardsScope Verify
Treat model output as untrusted decision support.
AI assistance remains subject to the same permission, scope, validation and output-handling expectations as the rest of the application.
What records is the model allowed to receive?
Use the signed-in user, organisation, page permission and dashboard context.
How is important output checked?
Return to source records and responsible professional judgement before action.
Provider credentials remain server-side and model output must be constrained, validated and escaped before display, storage or action.
Expected management outcomes
Make control principles visible without overstating assurance.
Discuss deployment, data and control requirements directly and preserve the boundary between application features and independently verified claims.
Clear access decisions
Explain role, permission and data-scope behaviour.
Traceable operations
Support audit review and diagnostics with stable metadata.
Responsible AI use
Keep context, validation and human verification explicit.
See the connected workflow
Review the controls behind the workflow you plan to use.
Select the roles, data scope, export, audit or AI questions you want covered in the walkthrough.