🤖 Entra Agent ID: The blueprint blast radius

Entra Agent ID is an extension of Entra's application model that provides identities for AI agents. Unlike applications, the agent identity model allows linking a single app registration (blueprint) to multiple identities and their associated privileges, increasing the potential blast radius of a compromised agent.

https://securitylabs.datadoghq.com/articles/agent-id-blueprint-blast-radius

#AI

🤖 AI Agent Supply-Chain Malware in Instruction Files

Mitiga Labs scanned 50,000+ AI instruction files across 7,000+ repos, finding prompt-exfiltration malware, ANTHROPIC_BASE_URL MITM overrides, YOLO-mode permission bypasses, and 1,230+ hardcoded API keys. They released Skillgate, a free scanner, to detect these techniques.

https://www.mitiga.io/blog/malware-in-ai-instruction-files-skillgate

#AI

👩‍💻 Holding blobs for ransom: Four methods for Azure Storage ransomware

This post explores four vectors for threat actors to abuse Azure Storage to maliciously encrypt victim blobs, including step-by-step explanations and event codes for detection.

https://securitylabs.datadoghq.com/articles/azure-blob-storage-ransomware-four-methods

#azure

🤖 An Introduction to AI Coding Agent Security

NCC Group whitepaper covering security of AI coding agents (Claude Code, Cursor, Codex) as of May 2026. Analyzes permission models, OS-level sandboxing, agent tool attack surfaces, and common vuln classes: workspace trust bypasses, sandbox escapes, permission prompt bypasses, sensitive file overwrites, and indirect prompt injection.

#AI

🤖 LLMjacking evolved: Attackers are using stolen AI compute to build offensive agentic tools

Sysdig TRT caught a threat actor using a stolen Ollama server to power an autonomous, multi-stage offensive hacking tool.

https://www.sysdig.com/blog/llmjacking-evolved-attackers-are-using-stolen-ai-compute-to-build-offensive-agentic-tools

#AI

🤖 datadog-saist

Unlike traditional SAST tools that rely solely on parsing and analysis rules, this project uses LLM (e.g. Claude from Anthropic, GPT from OpenAI or Gemini from Google) to find vulnerabilities.

https://github.com/DataDog/datadog-saist

#AI

🔐 Enterprise-Managed Authorization: Zero-touch OAuth for MCP

The Enterprise-Managed Authorization extension to the Model Context Protocol is now stable, enabling organizations to centrally provision MCP server access through their identity provider so users get connected servers on first login without per-app OAuth.

https://blog.modelcontextprotocol.io/posts/enterprise-managed-auth

#AIM

🔴 Securing agentic AI: What's new in VPC Service Controls

Designed for agentic workloads, new capabilities in VPC Service Controls can help establish a network-level, destination-based perimeter.

https://cloud.google.com/blog/products/identity-security/securing-agentic-ai-whats-new-in-vpc-service-controls

#gcp

🤖 AI brands as bait: How threat actors are using the AI hype in social engineering

Threat actors impersonate AI brands (ChatGPT, Claude, DeepSeek, Copilot) in phishing, malvertising, and SEO-poisoning campaigns to steal credentials, credit card data, and deliver infostealers like Vidar. Storm-3075 and Fox Tempest are attributed actors using signed malware.

https://www.microsoft.com/en-us/security/blog/2026/06/08/ai-brands-as-bait-how-threat-actors-are-using-the-ai-hype-in-social-engineering

#AI