Review a fictional role-permission matrix, identify excess access and prepare a traceable change proposal.
A workshop portal has Viewer, Editor and Approver roles. Viewers may read published files; Editors may also create drafts; Approvers may review drafts and publish. The current matrix accidentally gives Viewer the publish permission and gives Editor role-management permission.
| Role | Read published | Create draft | Review draft | Publish | Manage roles |
|---|---|---|---|---|---|
| Viewer | 1 | 0 | 0 | 1 | 0 |
| Editor | 1 | 1 | 0 | 0 | 1 |
| Approver | 1 | 0 | 1 | 1 | 0 |
Start from the operations each role must perform. Separate ordinary content actions from administration of roles themselves. Least privilege is a task-based design principle, not a reason to remove access without understanding the work. The case defines a simplified baseline for a fictional portal.
Write the intended matrix from the case narrative.
Required-access baseline
Record your result and evidence here or in a separate file.
No ordinary content role requires role management.
Tie each allowed permission to a named task.
Compare each role-action pair and classify it as required and granted, required but missing, or granted without a stated need. A matrix makes excess access visible but does not show per-resource conditions. Preserve the current state as evidence and propose changes in a separate table.
Find the two excess permissions and produce a difference report.
Access difference report
Record your result and evidence here or in a separate file.
Viewer/publish and Editor/manage roles are the two excess grants.
Compare every cell rather than only the most powerful role.
Role membership may be only one part of authorization. An editor might be allowed to create drafts but not edit every organizer’s draft. Record these conditions separately from the coarse matrix. The exercise does not inspect a live implementation, so conditions remain requirements to verify later.
Add ownership and approval-separation questions to the review.
Conditional-access questions
Record your result and evidence here or in a separate file.
A role label is not treated as proof of resource-level enforcement.
Name the resource and action in each question.
A change proposal should name the exact grant to remove, the reason, owner and rollback plan. Consider whether legitimate work depends on the access and how that work will continue. No actual account changes are needed for this learning case. Keep the proposed and approved states distinct.
Write two change requests with checks for required tasks after removal.
Permission change proposals
Record your result and evidence here or in a separate file.
Each proposal is specific and preserves the required task set.
Do not use remove all access as a substitute for a targeted correction.
A review record should show who assessed access, when, against which baseline and with what evidence. Add a review trigger for role changes, departures or new sensitive actions. The package provides the template only; it does not create an external schedule or modify any user account.
Build the review register and a completion checklist.
Access-review handoff
Record your result and evidence here or in a separate file.
No proposed removal is recorded as already applied.
Include status and verification fields for each change.
Deliver the baseline, current-versus-required comparison, conditional-access questions, change proposals and review register. Add a missing required permission in a copy and show how the checker distinguishes it from excess access.
The report preserves the current state and does not claim real permission changes or verified enforcement.
| Criterion | Points | Evidence |
|---|---|---|
| Evidence and correctness | 30 | Calculations, source use and results match the supplied case; limitations are honest. |
| Required artifacts | 30 | All five stage outputs and the capstone handoff are present and inspectable. |
| Reasoning and decisions | 20 | Choices, assumptions, alternatives and the case-specific critical condition are explained. |
| Reproducibility | 10 | A reviewer can follow the stated steps and distinguish expected from observed results. |
| Clear communication | 10 | Labels, units, structure and evidence references make the work easy to review. |
Score at least 80/100, supply every required artifact, and satisfy the path-specific critical condition. Rework missing evidence before marking complete.
The current matrix has two grants not present in the case baseline: Viewer can publish and Editor can manage roles. Removing those two in the proposed matrix leaves Viewer read-only, Editor able to read and create drafts, and Approver able to read, review and publish.
This is a fictional authorization review, not identity-provider administration or a completed security audit.