Josh Botz | AI Agent Security

I work in cloud security at Ontic, a regulated SaaS company. My day-to-day covers FedRAMP, continuous monitoring, and GRC.

This channel is for security leaders, architects, and practitioners working through AI agent security. The shift to autonomous agents - systems that call tools, modify records, and act on a user's behalf - is changing what controls need to look like. Identity, acceptable use, kill-switches, monitoring: most of iit was designed for humans.
  
What this channel covers:
  • Lab walkthroughs reproducing attacks from new research
  • Breakdowns of recent CVEs and disclosures affecting AI agents
  • Practical control patterns (AUPs, identity, kill-switch design) you can take back to your team
  • Notes on the MCP supply chain and what to watch for

Lab work and code: github.com/joshbotz
LinkedIn: linkedin.com/in/joshbotz

Husband. Father. Follower of Christ.