Over 10 years we help companies reach their financial and branding goals. Engitech is a values-driven technology agency dedicated.

Gallery

Contacts

Via Giosuè Carducci, 21 - Pomigliano d'Arco (Italy)
Paseo Montjuic, número 30 - Barcelona (Spain)

info@hacktivesecurity.com

+39 06 8773 8747

AI Application Security

datapizza-ai, Yet Another Vulnerable AI Framework

TL;DR Two Remote Code Execution (RCE) vulnerabilities were identified in datapizza-ai framework: What is datapizza-ai Source here https://github.com/datapizza-labs/datapizza-ai. CVE-2026-2969 The vulnerability is caused by the usage of vulnerable functions of Jinja2 template engine (datapizza-ai-core/datapizza/modules/prompt/prompt.py, source here https://github.com/datapizza-labs/datapizza-ai/blob/v0.0.2/datapizza-ai-core/datapizza/modules/prompt/prompt.py). To reproduce the exploit we have to install datapizza-ai: Create a python file with the following content: Execute […]

AI Application Security

CVE-2025-67511: Tricking a Security AI Agent Into Pwning Itself

TL;DR A command injection vulnerability in the function tool run_ssh_command_with_credentials() available to AI agents in cai-framework <= 0.5.9 allows to achieve Remote Command Execution. This post describes a vulnerability I discovered in the CAI framework in which incomplete shell escaping inside a tool accessible to AI agents allowed hostile targets to trigger command execution on […]