Factory launches Automated Security Review in Droid

Embedding security threat modeling directly into AI developer agents shifts security reviews from a post-development gatekeeper to a continuous, real-time participant in the development lifecycle.

• –Structured methodology: Utilizing the STRIDE framework ensures that coding agents cover a comprehensive range of threats, including privilege escalation and information disclosure.
• –Interactive audits: The ability to invoke audits on-demand via slash commands enables developers to secure their work prior to opening a PR.
• –Production validation: Successfully identifying and disclosing a WorkOS Node SDK vulnerability validates the tool's real-world bug-hunting capability.