Today, an AI agent is often the one writing the code, pulling the package, and opening the pull request. The software supply chain most teams built was never designed for autonomous actors moving that fast, and speed without guardrails is just risk moving quicker.

In this session, JFrog and AWS show how to give an AI agent enough autonomy to be useful and enough governance to be trusted. Live, Eldad Assis (JFrog) and Arik Yakir (AWS) follow one artifact from build to production, with the focus on Amazon Bedrock AgentCore Gateway: the secure entry point that governs the tools and dependencies a production agent can reach. JFrog Curation, JFrog Artifactory, and JFrog Xray keep chain-of-custody intact the whole way, and Kiro shows where it starts in the IDE.

What you'll take away:

• Where agentic development adds risk, and where it removes friction if done right
• How Amazon Bedrock AgentCore Gateway governs the tools and dependencies a production agent can reach
• How JFrog keeps every artifact trusted, from the IDE to the running agent

Speakers: Eldad Assis, Principal DevOps Architect, CTO Office, JFrog · Arik Yakir, Senior Partner Solutions Architect, AWS