Posted on
in Presentations
AI agents can ingest or leak sensitive data through MCP servers. This talk will show how to build a small MCP sentinel that scans requests and tool arguments, detects partial and transformed sensitive content, and blocks unsafe use with local audit trails. An example moves from the clipboard to the agent, gets intercepted, and is refused. Attendees take home code and a checklist to deploy quickly.
Access This and Other RSAC™ Conference Presentations with Your Free RSAC Membership
Your RSAC™ Membership also includes AI-powered summaries, mind maps, and slides for Conference presentations, Group Discussions with experts, and more.
Watch Now >>
Share With Your Community