stakpak

You've seen AI agents deleting databases and wiping codebases. Whitelists and blacklists fail because LLMs can do the same destructive thing in 100 different ways. @kajogo777 from @stakpak_dev has built novel deterministic security guardrails that make agents safe for production. At AI Native DevCon, he'll introduce Warden—a new security enforcer that creates boundaries agents cannot cross, regardless of the tools they use. The demo covers secure secret handling, mTLS for MCP, and real examples from Stakpak's open-source DevOps agent. Reference implementation in Rust included. This is your first look at techniques that make coding agents production-safe. 📍 DevCon is November 18-19th in NYC. 📷 Tickets and full lineup: ainativedev.co/vtf Use the discount code GEORGE25 to get 25% off your tickets