AWS Security Agent adds threat modeling, Kiro power and Claude Code plugin, and more

AWS previewed new AWS Security Agent capabilities.

It can now generate threat models from design docs or code and improve PR/repository scanning with remediation and validation.

Integrations now extend to GitHub, GitLab, Bitbucket, and Confluence, with MCP support for Kiro, Claude Code, and other AI IDEs.

The update moves security earlier in the development workflow, helping teams find issues faster and reduce manual bottlenecks.