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.
AI-based security is used to describe everything from a genuinely trained detection model to a marketing rewrite of a rules engine. Here's how to tell what you're actually buying.
Direct and indirect prompt injection are different attack vectors with different blast radii. Real 2025 CVEs like EchoLeak show why the distinction matters.
LLM jailbreaking bypasses AI safety guardrails through techniques like DAN prompts, Crescendo, and Skeleton Key — here's how it works and how to defend against it.
Model weights are executable artifacts, not data. Here's how AI model supply chain attacks work, from pickle exploits to weight tampering, and how to stop them.
A vulnerable transitive dependency may require upgrading an ancestor. Griffin AI computes the cascade; Mythos-class tools often stop at the first level.
Retrieval-augmented generation was the 2024 success story. 2026 is when RAG poisoning moved from research to production incidents.
MCP security explained: how tool poisoning, rug pulls, and 2025's critical CVEs (mcp-remote, MCP Inspector) put AI agents at risk—and how to defend against them.
AI agent security explained: how autonomous AI agents get attacked through prompt injection, tool poisoning, and exposed MCP servers, and how to stop it.
AI coding assistants now write nearly half of some codebases—and research shows 45% of that code ships with exploitable flaws. Here's what security teams need to know.
Weekly insights on software supply chain security, delivered to your inbox.
Your first fix PR is minutes away.
No sales call required, even your agent can complete the purchase over MCP.
