Research & Blog
Research from Variablee, the frontier AI security lab behind Variablee Code
White paper

Route-Sliced Context: Scalable LLM-Based Detection of Broken Object-Level Authorization in Large Codebases

Variablee research team · November 2026

The core algorithm behind Variablee Code: why BOLA is a route-local property, how we build a bounded context slice for every route across nine frameworks, and how it performs on 25 documented vulnerabilities (24 detected, one false positive on 28 known-safe routes).

Read the white paper →
Blog

AI-native SAST for IDOR / BOLA

Sella Rafaeli, Security Research Lead · November 9, 2026

Why AI code review runs out of context on large codebases, how narrowing it to one route at a time solves that, and how we benchmark the scanner on applications with known, published vulnerabilities.

Read the post →
Contact us

Talk to our team

Want to run Variablee Code on your organization's codebase, plan a pilot, or ask about our research? Send us a note and we'll get back to you.

  • Pilots on large enterprise codebases
  • Private repositories and SSO
  • Research collaborations

We use what you send only to reply to you. See our Privacy Policy.