Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore advanced deception engineering techniques in this conference talk that demonstrates building "MCP Threat Trap," a sophisticated honeypot designed for Zero Trust environments and AI-driven threat detection. Learn to construct honeypots that simulate sensitive internal tools like Okta admin password resets using the MCP protocol, complete with realistic delays, secure error handling, and SSE streams that authentically mimic enterprise APIs. Discover how to implement silent Canarytoken triggers that capture comprehensive metadata including user agents, IP addresses, and sensitive account attempts without alerting intruders. Master serverless deployment strategies using Cloudflare Workers for global edge distribution, eliminating the need for EC2 instances, patching, or infrastructure management while maintaining stealth and instant scalability. Examine real-world attack data from unsolicited hits including abuse-flagged cloud scanners and human reconnaissance attempts, and understand validation techniques using OWASP AI Security scenarios to catch AI-agent driven reconnaissance. Gain insights into transforming random scans into actionable intelligence that feeds Zero Trust policies and provides incident response teams with context-rich alerts. Explore concepts for evolving deception surfaces into adaptive systems that dynamically respond to attacker interactions, and acquire a repeatable blueprint for deploying globally distributed honeypots while learning critical implementation pitfalls to avoid.
