CVE-2025-3579: Aidex command execution through prompt injection in /api/<string-chat>/message

Summary

CVE-2025-3579 affects Aidex versions prior to 1.7. An authenticated user, taking advantage of an open registry, can execute unauthorised commands on the system, including Unix OS commands and calls to internal services such as PHP or MySQL, and native functions of the Laravel or Symfony framework. The attack is carried out through Prompt Injection via the /api/<string-chat>/message endpoint by manipulating the 'content' parameter.

Mitigation

The source does not state a fix yet. Check the original advisory for updates.

Related items

• InfoTop agent skill security resources — November 2026Similar attack · Adversa AI Blog
• InfoHow to Turn AI Red Team Findings Into Runtime ControlsSimilar attack · Straiker Blog
• MediumDocker Sandboxes vulnerability lets AI agents access sensitive files outside their sandbox (CVE-2026-103916) | Endor LabsSimilar attack · Endor Labs Blog
• MediumFree the Model, Sell the Gateway: Prompt Adware hidden in Jailbreak BundlesSimilar attack · Check Point Research
• InfoRight Tool, Wrong Parameters: Defending LLM Agents Against Parameter PoisoningSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)