Model Context Protocol (MCP) use is increasing in popularity for connecting AI agents to data sources, and other services. But so too are vulnerabilities that bring unique risks to agentic systems.

Microsoft has observed two named Chinese nation-state actors, Linen Typhoon and Violet Typhoon, exploiting vulnerabilities ...

Malicious GitHub repositories used by threat actors to host Amadey payloads and steal data, impacting targeted entities.

npm packages hit by phishing-based supply chain attack, exposing developers to malware and remote access threats.

Microsoft issues an apology for "inconvenience or confusion" caused after it mistakenly indicated that it had resolved a ...

If a website needs you to log in, you can do that securely through a special browser view, which lets the agent dig deeper ...

OpenAI launched ChatGPT Agent, a new feature that lets the company's AI assistant complete multi-step tasks by controlling ...

The stolen API key allows interaction with at least 52 large language AI models (LLMs) that are in development at Musk's xAI, ...

Initial community reactions to Kiro were mixed, but developers were intrigued, praising the emphasis on specs, hooks and structure.

This increase in AI usage has led to a distinction between officially sanctioned, well-governed AI tools (safe AI) and ...

Many open-source repositories contain privileged GitHub Actions workflows that execute untrusted code and can be triggered by attackers to expose credentials and access tokens, as MITRE and Splunk ...