prompt-injection
Safeguard articles tagged "prompt-injection" — guides, analysis, and best practices for software supply chain and application security.
127 articles
Agentic AI Security Took Center Stage: The OWASP GenAI Summit at Infosecurity Europe 2026
OWASP's first dedicated GenAI Security Summit at Infosecurity Europe put agentic AI security front and center, unveiling an Agentic Research Council and a maturity framework. Here's what actually mattered.
Cursor's AI security agents: what they get right and what's missing
Cursor's Bugbot and MCP agents catch real bugs, but CurXecute and MCPoison show they open new attack surfaces SCA tools never had to face.
Securing AI coding IDE extensions and plugins
VS Code themes with 9M installs shipped backdoors; Cursor's rules files were hijacked in 2025. Here's what AI IDE extension security actually requires.
LLM output validation and sanitization best practices
LLM output is untrusted input to everything downstream. Here's how to validate, encode, and sandbox it before it becomes your next injection vector.
Agentic AI Security: Access Control Is the Whole Ballgame
Gartner expects most successful attacks on AI agents through 2029 to exploit access control, with prompt injection as the delivery mechanism. Here is why that single failure mode dominates agentic AI security, and what actually moves the needle.
Cisco's May 2026 Multi-Turn Jailbreak Study: Why Frontier Model Safety Collapses Over a Conversation
Cisco's AI threat team tested 15 flagship models with ~7,000 multi-turn attacks and found success rates as high as 88 percent. Single-turn safety scores told defenders almost nothing about real-world resilience.
Claude Opus 4.8 for Security Teams: Capabilities, AppSec Use, and Governance (May 2026)
Anthropic shipped Claude Opus 4.8 on May 28, 2026, with sharper agentic coding and better honesty about its own work. Here is what it changes for vulnerability triage, fix-PRs, and the governance you need before it touches your pipeline.
Cursor MCP Security: The Risks and How to Harden It
Cursor's MCP support lets the AI editor call external tools and data sources. That power comes with real risks. Here is how Cursor MCP can be attacked and hardened.
Hacking AI: How Attackers Target Machine Learning Systems
Hacking AI is not science fiction; it is a growing set of concrete techniques that exploit how models learn, process input, and produce output. Here is how to think about defending against them.
Prompt-injection vectors specific to MCP servers and how to layer defenses
MCP servers expose three distinct prompt-injection surfaces — resource contents, tool outputs, and sampling requests — and each one needs its own defense layer. Here is how to think about them together.
MCP tool poisoning: hidden instructions and rug-pulled tool definitions
Tool-poisoning attacks against Model Context Protocol servers hide adversarial instructions inside tool descriptions and silently mutate tool definitions after install. Here is how the attack works and how to defend against it.
Prompt injection attacks against AI coding/security tools
AI coding assistants like Copilot and Cursor can be hijacked by hidden text in files, comments, and packages. Here's how prompt injection malware works and how Safeguard detects it.