Multi-Step IDOR Exploitation
What you'll learn: How attackers chain multiple requests together to exploit IDOR vulnerabilities by manipulating object references through different application states or workflows.
The Core Concept
Many applications don't expose objects directly in a single request. Instead, you must complete multiple steps—like creating a draft, then editing it, then publishing it. Multi-step IDOR exploitation means an attacker chains these requests together, manipulating object IDs at each stage to access or modify resources they shouldn't touch.
Think of it like a hotel: you can't just walk into any room. First you check in (step 1), receive a key card (step 2), then access your room (step 3). But what if you could swap someone else's room number into step 3 while using your legitimate key from step 2?
How It Works
The attacker identifies a workflow with multiple stages:
- Initial creation – Start a legitimate process (create order #123)
- State transition – Move through intermediate states (order moves to "pending")
- Final action – Exploit the final step by referencing someone else's object ID
For example:
- POST
/api/invoicescreates invoiceinv_456(yours) - GET
/api/invoices/inv_456/reviewshows review page - POST
/api/invoices/inv_789/approve– attacker changes ID to approve someone else's invoice
Each step might check permissions independently (or not at all), creating gaps where authorization isn't consistently enforced throughout the entire workflow.
Why This Matters
Developers often secure the "entry points" but forget to validate ownership at every state transition. An attacker with access to the workflow can substitute IDs mid-process, especially when state is tracked server-side but object references come from client input.
Key Takeaway: Multi-step IDOR attacks exploit inconsistent authorization across workflow stages—always validate object ownership at every step, not just the first one.