AI Security
vLLM CVE-2025-62164: Tensor Deserialization RCE
vLLM 0.10.2-0.11.0 deserialized user-supplied PyTorch tensors via torch.load() in the Completions API. Memory corruption, potential RCE.
Jul 26, 20265 min read
Resources
Supply Chain Security, in plain English.
Deep dives, practical guides, and incident analyses from engineers who build Safeguard. No fluff, no vendor FUD — just what you need to ship secure software.
Filter
Articles
RSS feedCompliance
DORA Concentration Risk: ESAs' Designation of Critical ICT Third-Party Providers
DORA Article 31 lets the ESAs designate critical ICT third-party providers (CTPPs) for direct EU-level oversight. First designations land in 2025-2026 from the Register of Information.
Jul 26, 20268 min read
AI Security
Claude Opus 4.5 System Card: Defender Takeaways
Anthropic released Claude Opus 4.5 on November 24, 2025 with the most detailed safety section of any system card to date. We pull out what enterprise defenders should change.
Jul 26, 20267 min read
DevSecOps
Jenkins Stapler Unauthenticated RCE Mass-Exploited for Cr...
CVE-2018-1000861 let attackers hit Jenkins Stapler unauthenticated, planting cryptomining malware on exposed CI/CD build servers across the internet.
Jul 25, 20267 min read
DevSecOps
Jenkins CLI Java Deserialization Remote Code Execution (C...
CVE-2017-1000353 let attackers gain unauthenticated RCE on Jenkins via CLI Java deserialization. Here's the impact, timeline, and how to remediate it.
Jul 25, 20268 min read
DevSecOps
Jenkins Remote Code Execution via Groovy Metaclass (CVE-2...
CVE-2016-0792 let attackers bypass Jenkins' deserialization blacklist using Groovy's metaclass to achieve unauthenticated remote code execution via the CLI.
Jul 25, 20269 min read
DevSecOps
Jenkins CLI Deserialization RCE via Commons-Collections G...
CVE-2015-8103: unauthenticated RCE in Jenkins CLI via a Commons-Collections deserialization gadget chain. Impact, timeline, and remediation.
Jul 25, 20269 min read
AI Security
AI Model Watermarking and Provenance
Watermarking and provenance are the two most confused terms in AI security. A practical breakdown of what each actually does, where the 2025 techniques break, and what to ship in the meantime.
Jul 25, 20267 min read
AppSec
Code Security Scan: How to Scan Your Code for Vulnerabilities
A code security scan analyzes your source and its dependencies for security flaws before they ship. Here is how the main scan types work, what tools to use, and how to wire scanning into CI without drowning in noise.
Jul 25, 20267 min read
Stay informed
Weekly insights on software supply chain security, delivered to your inbox.
Self-healing security runs on Safeguard.
Your first fix PR is minutes away.
No sales call required, even your agent can complete the purchase over MCP.