Blind IDOR and Indirect Object References
What you'll learn: How to detect and exploit IDOR vulnerabilities when the application doesn't directly show you whether your attack succeeded.
What Is Blind IDOR?
In traditional IDOR attacks, you immediately see the results—change a user ID and you see someone else's profile. Blind IDOR is trickier: the application doesn't give you direct feedback about whether you accessed someone else's data. The response looks the same whether you succeeded or failed.
Think of it like sending letters to addresses you've guessed. In normal IDOR, you see the letter's contents immediately. With blind IDOR, you only get a "message sent" confirmation—you won't know if it reached the right person until you check elsewhere (email notifications, audit logs, or side effects).
Indirect Object References
These vulnerabilities exploit mappings between user-facing identifiers and internal resources. Instead of manipulating userId=123, you might tamper with:
reportId=5which internally maps touserId=456invoiceNumber=ABC-789which references another customerfileToken=xyz123which unlocks arbitrary documents
The application translates your input into backend references without proper authorization checks.
Detection Techniques
Since feedback is absent, you must look for:
- Side effects: Changed settings, triggered emails, modified timestamps
- Timing differences: Successful operations may take different amounts of time
- Out-of-band callbacks: Webhooks or notifications sent to attacker-controlled endpoints
- Secondary channels: Check if actions appear in activity logs or notifications elsewhere
For example, updating preferences?accountId=999 might not show you the data, but you could verify success by logging in as that account or checking if a confirmation email was sent.
Key Takeaway: Blind IDOR requires creative detection through side effects and indirect evidence, while indirect object reference attacks exploit insufficient authorization on internal mappings between user inputs and backend resources.