[Security]
Security
Threat analysis and practical guidance for securing AI models, agents, MCP servers, and the tools your team ships with.
MCP Security Best Practices: 6 Attack Vectors Every Team Should Know
The MCP specification defines six critical attack vectors, from confused deputy exploits to session hijacking. Here's how each one works and what to do about it.
Stanislas HannebelleApril 11, 2026How to Build a Secure AI Agent: A Practical Guide for Startups
AI agents are shipping fast, but most teams underestimate the security implications. This guide covers the real threats, from prompt injection to output exfiltration, and gives you actionable patterns to build agents that fail safely.
Arnaud DrizardApril 11, 2026ShadowPrompt: How a Zero-Click Vulnerability in Claude's Chrome Extension Could Hijack Your Browser
A zero-click vulnerability in Anthropic's Claude Chrome extension allowed any website to silently inject prompts and steal sensitive data. Here's what happened, how it worked, and what it means for your AI tool governance.
Arnaud DrizardMarch 31, 2026We Built a Customer-Facing MCP Server. Here's What the Spec Didn't Prepare Us For.
Building a customer-facing MCP server? Here's what the spec misses: OAuth IdP gaps, client divergence, multi-tenant auth, and supply chain risk.
Stanislas HannebelleMarch 3, 2026AI Coding Assistants Are Now a Security Risk: What SOC 2 and ISO 27001 Companies Need to Know
From malicious extensions exfiltrating code to prompt injection attacks enabling remote execution, AI coding tools introduce new risks that most compliance frameworks don't explicitly address. Here's what CTOs and CISOs need to implement now.
Arnaud DrizardFebruary 13, 2026[Get started]
Secure the agent era
See, secure, and govern every AI model, agent, and MCP server from one transparent proxy.
Book a demo