Track vulnerabilities like this one. Sign up free to get alerted when software you use is affected.
9.8

Open GenAI Stack can run unwanted code

CVE-2026-77177 · published 20 days ago
Summary

The Open GenAI Stack used in Meta AI services can be tricked into executing code through specially crafted prompts. This could let an attacker run commands on the server, potentially compromising data or system integrity. Apply the vendor's security update or disable template processing until a fix is applied.

Original advisory text
Open GenAI Stack (aka ogx-ai) 2026-06-11, as used in the Meta AI backend for WhatsApp and other products, allows code execution because prompt injection (with Jinja2 template syntax) can be used to...
Open GenAI Stack (aka ogx-ai) 2026-06-11, as used in the Meta AI backend for WhatsApp and other products, allows code execution because prompt injection (with Jinja2 template syntax) can be used to achieve server-side expression evaluation without sanitization.
References
Fix within
Internet-facing 3 days
Internal 60 days
  • Not known to be exploited
  • Can be exploited automatically
  • Gives an attacker full control
Severity
9.8 Critical
Exploitation
<1% chance of attack within 30 days
Type
CWE-94Code Injection
Timeline
Published29 Sep 2026
Updated15 Oct 2026
First seen29 Sep 2026
Sources
CVE-2026-77177 · NVD
CVE-2026-77177 · MITRE
Track software like this
Free during beta