| name | django-access-review |
| description | Reviews Django and DRF access control by tracing whether User A can read, change, or delete User B objects: get_queryset scoping, permission_classes, has_object_permission, tenant managers, and IDOR on pk routes. Use when auditing Django authorization, object permissions, or tenant isolation. Not for Flask/FastAPI auth, generic SAST pattern dumps without a traced flow, or comment-only fixes. |
| version | 1.0.1 |
Overview
Find access control vulnerabilities by investigating how the codebase answers one question: Can User A access, modify, or delete User B's data?
This skill uses an investigation-driven approach rather than generic pattern matching. Every codebase implements authorization differently; your job is to understand the specific implementation, then find gaps.
When to Use
- You need to review Django or DRF code for access control gaps, IDOR risk, or object-level authorization failures.
- The task involves confirming whether one user can access, modify, or delete another user's data.
- You want an investigation-driven authorization review instead of generic pattern matching.
Prerequisites
- Access to the target Django/DRF codebase.
- PowerShell environment if running search commands on a Windows host.
Procedure
Phase 1: Understand the Authorization Model
Before looking for bugs, answer these questions about the codebase:
- Where are permission checks implemented?
- Decorators? (
@login_required, @permission_required, custom?)
- Middleware? (
TenantMiddleware, AuthorizationMiddleware?)
- Base classes? (
BaseAPIView, TenantScopedViewSet?)
- Permission classes? (DRF
permission_classes?)
- Custom mixins? (
OwnershipMixin, TenantMixin?)
- How are queries scoped?
- Custom managers? (
TenantManager, UserScopedManager?)
get_queryset() overrides?
- Middleware that sets query context?
- What's the ownership model?
- Single user ownership? (
document.owner_id)
- Organization/tenant ownership? (
document.organization_id)
- Hierarchical? (org -> team -> user -> resource)
- Role-based within context? (org admin vs member)
Investigation commands (PowerShell):
# Find how auth is typically done
Get-ChildItem -Recurse -Filter *.py | Select-String -Pattern "permission_classes|@login_required|@permission_required" | Select-Object -First 20
# Find base classes that views inherit from
Get-ChildItem -Recurse -Filter *.py | Select-String -Pattern "class Base.*View|class.*Mixin.*:" | Select-Object -First 20
# Find custom managers
Get-ChildItem -Recurse -Filter *.py | Select-String -Pattern "class.*Manager|def get_queryset" | Select-Object -First 20
# Find ownership fields on models
Get-ChildItem -Recurse -Filter models.py | Select-String -Pattern "owner|user_id|organization|tenant" | Select-Object -First 30
Do not proceed until you understand the authorization model.
Phase 2: Map the Attack Surface
Identify endpoints that handle user-specific data:
- What resources exist?
- What models contain user data?
- Which have ownership fields (
owner_id, user_id, organization_id)?
- Which are accessed via ID in URLs or request bodies?
- What operations are exposed?
For each resource, map:
- List endpoints: what data is returned?
- Detail/retrieve endpoints: how is the object fetched?
- Create endpoints: who sets the owner?
- Update endpoints: can users modify others' data?
- Delete endpoints: can users delete others' data?
- Custom actions: what do they access?
Phase 3: Ask Questions and Investigate
For each endpoint that handles user data, ask: "If I'm User A and I know the ID of User B's resource, can I access it?"
Trace the code to answer this:
- Where does the resource ID enter the system? (URL path, query param, request body)
- Where is that ID used to fetch data? (Find the ORM query or database call)
- Between (1) and (2), what checks exist?
- Is the query scoped to current user?
- Is there an explicit ownership check?
- Is there a permission check on the object?
- Does a base class or mixin enforce access?
- If you can't find a check, is there one you missed? (Check parent classes, middleware, managers, URL-level decorators)
Follow-Up Questions:
- For list endpoints: Does the query filter to user's data, or return everything?
- For create endpoints: Who sets the owner - the server or the request?
- For bulk operations: Are they scoped to user's data?
- For related resources: If I can access a document, can I access its comments? What if the document belongs to someone else?
- For tenant/org resources: Can User in Org A access Org B's data by changing the
org_id in the URL?
Phase 4: Trace Specific Flows
Pick a concrete endpoint and trace it completely.
Example Investigation:
- Find the view handling
GET /api/documents/{pk}/ -> DocumentViewSet.retrieve() in api/views.py
- Check what
DocumentViewSet inherits from -> class DocumentViewSet(viewsets.ModelViewSet) (No custom base class with authorization)
- Check
permission_classes -> permission_classes = [IsAuthenticated] (Only checks login, not ownership)
- Check
get_queryset() -> return Document.objects.all() (Returns ALL documents!)
- Check for
has_object_permission() -> Not implemented
- Check
retrieve() method -> Uses default, which calls get_object() -> get_object() uses get_queryset(), which returns all
- Conclusion: IDOR - Any authenticated user can access any document
What to look for when tracing:
- Potential gap indicators (investigate further, don't auto-flag):
get_queryset() returns .all() or filters without user
- Direct
Model.objects.get(pk=pk) without ownership in query
- ID comes from request body for sensitive operations
- Permission class checks auth but not ownership
- No
has_object_permission() and queryset isn't scoped
- Likely safe patterns (but verify the implementation):
get_queryset() filters by request.user or user's org
- Custom permission class with
has_object_permission()
- Base class that enforces scoping
- Manager that auto-filters
Phase 5: Report Findings
Only report issues you've confirmed through investigation.
Confidence Levels:
- HIGH: Traced the flow, confirmed no check exists. Report with evidence.
- MEDIUM: Check may exist but couldn't confirm. Note for manual verification.
- LOW: Theoretical, likely mitigated. Do not report.
Suggested Fixes Must Enforce, Not Document:
A comment or docstring does not enforce authorization. Your suggested fix must include actual code that:
- Validates the user has permission before proceeding
- Raises an exception or returns an error if unauthorized
- Makes unauthorized access impossible, not just discouraged
Example of a BAD fix suggestion:
def get_resource(resource_id):
return Resource.objects.get(pk=resource_id)
Example of a GOOD fix suggestion:
def get_resource(resource_id, user):
resource = Resource.objects.get(pk=resource_id)
if resource.owner_id != user.id:
raise PermissionDenied("Access denied")
return resource
Report Format:
## Access Control Review: [Component]
### Authorization Model
[Brief description of how this codebase handles authorization]
### Findings
#### [IDOR-001] [Title] (Severity: High/Medium)
- **Location**: `path/to/file.py:123`
- **Confidence**: High - confirmed through code tracing
- **The Question**: Can User A access User B's documents?
- **Investigation**:
1. Traced GET /api/documents/{pk}/ to DocumentViewSet
2. Checked get_queryset() - returns Document.objects.all()
3. Checked permission_classes - only IsAuthenticated
4. Checked for has_object_permission() - not implemented
5. Verified no relevant middleware or base class checks
- **Evidence**: [Code snippet showing the gap]
- **Impact**: Any authenticated user can read any document by ID
- **Suggested Fix**: [Code that enforces authorization - NOT a comment]
### Needs Manual Verification
[Issues where authorization exists but couldn't confirm effectiveness]
### Areas Not Reviewed
[Endpoints or flows not covered in this review]
Pitfalls
- Suggesting documentation as a fix: Comments and docstrings do not enforce authorization. Always suggest actual code that validates permissions and raises exceptions.
- Auto-flagging patterns: Do not scan for predefined vulnerable patterns without tracing the actual data flow. A
.all() query might be safe if a middleware restricts it.
- Missing hidden checks: Authorization might be enforced in parent classes, middleware, or custom managers. Do not conclude a vulnerability exists without checking these layers.
- Ignoring ownership assignment: For create endpoints, verify whether the owner is set server-side or if it can be manipulated via the request body.
Verification
Use this checklist to guide your review, not as a pass/fail checklist: