ThreatsCloud & SaaSCritical
GitLab AI Gateway Prompt Execution Flaw Exposes Cloud Workspaces to Remote Code Execution (CVE-2026-90970)
CST Editorial Desk
Topic
AI Agent Security: news, advisories and analysis
4 articles
CyberSecureToday covers AI Agent Security across 4 articles published between 20 Sep 2026 and 1 Oct 2026. The coverage includes 4 vulnerability advisories. 4 advisories are rated critical or high severity.
Stories on AI Agent Security most often intersect with Model Context Protocol, Remote Code Execution and MCP, with Technology, Cloud & SaaS the most affected sector. The most recent story, “GitLab AI Gateway Prompt Execution Flaw Exposes Cloud Workspaces to Remote Code Execution (CVE-2026-90970)”, was published 1 Oct 2026. Every article is listed below, newest first, with the affected products, severity and remediation guidance in each piece.
Latest coverage
ThreatsCloud & SaaSHighMCP-Fetch Tool Flaw Exposes AI Agent Infrastructure to IPv6 SSRF Attacks
Daniel Okafor
ThreatsCloud & SaaSCriticalLiteLLM MCP Server Creation Flaw CVE-2026-30623 Enables Remote Code Execution via Malicious Tool Configurations
Mei-Lin Chen
ThreatsCloud & SaaSHighAmazon DynamoDB Model Context Protocol Server Code Injection Flaw Enables Arbitrary Host Execution
CST Editorial Desk