← Lab /Security & privacy

Brotli decoder out-of-bounds write enables remote code execution in popular runtimes

Security & privacy Disclosure · Prototype · Wiz Research

The claim

The authors assert that CVE-2026-111497 allows an unauthenticated remote attacker to achieve remote code execution by exploiting an out-of-bounds write in the Brotli decoder.

Merav Bar · Wiz Research Writeup

Published as: “Broccoli2shell: Everything You Need To Know About The Critical Brotli Decoder Vulnerability”

So what

The vulnerability puts the majority of cloud-hosted services at risk of remote code execution, making rapid patching of Brotli components a critical priority.

The five-minute version

The writeup describes CVE-2026-111497, an out-of-bounds write bug in the Brotli C decoder that can be triggered by malicious compressed data. Researchers demonstrated a proof-of-concept exploit against unpatched Node.js, showing that the flaw can lead to memory corruption, data disclosure, crashes, or remote code execution. Wiz telemetry indicates that 91% of cloud environments contain vulnerable instances, highlighting a broad attack surface. The authors provide patched version references and mitigation steps, but the exploitability of other listed products remains unverified.

What they found 4

What undercuts it the paper’s own results

Stated limits 2

What’s new here 4

Artifacts & methods 4

Topics

out-of-bounds writeBrotli compressionremote code executionmemory corruptioncloud vulnerabilitydependency managementpatchingcontainer security
← Previous Gheist’s GitHub token abuse blocked in public repos, still works in private/enterprise