Claude Security Adds Continuous Rescanning, Threat Models, and MCP Connector

Key Info

Claude Security is now in public beta on Claude Enterprise, letting teams connect repositories once and have Claude run a deep initial scan and then re-scan as code changes so findings track the current state of the code.

Highlights

• When a fix ships, Claude checks the changed code and marks the finding fixed, even if the team fixed it outside Claude Security; regressions reopen the finding.
• Each project gets a threat model that Claude writes from the code, which teams can edit.
• Projects can span repositories from GitHub, GitLab, and other Git providers.
• Findings reach other tools through a new API and MCP connector.
• Eligible orgs get $25K in usage credits per month for two months, through Dec 31, 2026.