Security3 dSeptember 18, 2026, 04:38

Finding the Agent Infrastructure: What Internet Measurement Can and Cannot Say About AI Coding Tool Exposure

In September 2026, a class of vulnerabilities CVE-2026-19592 was disclosed in AI coding agents including Claude Code, Codex, Goose, Qwen Code, and Grok Build. A malicious repository can influence subprocess calls and lead to command execution on the developer's machine.

DEV Community Open source

Finding the Agent Infrastructure: What Internet Measurement Can and Cannot Say About AI Coding Tool Exposure

Vulnerable: Claude Code, Codex, Goose, Qwen Code, Grok Build
Tracked as CVE-2026-19592
Attack is local, not visible to external scanning
Risk affects credentials to CI, registries, and clouds

#Claude#OpenAI#Qwen#Grok

Cyberattack hits University of Munich, student data at risk

Security

TASK#STOMP Windows backdoor steals documents, Wi-Fi passwords and screenshots

Security

CVE-2026-81657 in IBM Guardium: deserialization flaw rated 9.8

Security

Finding the Agent Infrastructure: What Internet Measurement Can and Cannot Say About AI Coding Tool Exposure

Cyberattack hits University of Munich, student data at risk

TASK#STOMP Windows backdoor steals documents, Wi-Fi passwords and screenshots

CVE-2026-81657 in IBM Guardium: deserialization flaw rated 9.8

ShinyHunters hijacks Cl0p ransomware site, demands extortion payment