Boston – November 04, 2026 -- Assail, Inc. has added Cassandra, a proactive AI agent, and expanded Threat Models to its Ares offensive security platform, linking attack scenarios to autonomous testing and remediation verification.
Cassandra monitors attack surface changes and drafts retest plans
Cassandra draws on existing assessment findings, asset information and discovery results. It flags newly exposed APIs, critical vulnerabilities, previously remediated issues that have resurfaced, and newly discovered assets. It prepares assessment and retest plans and coordinates follow-up through Ares, email or Slack. It also respects operator-defined restrictions such as excluded targets and deferred testing.
Three operating modes require human approval for every assessment
Organizations can choose Brief Only, Draft for Me or Act on Approval. Every assessment initiated through Cassandra requires human approval and stays within authorization, permissions and scope controls. The agent separates completed assessments from interrupted or failed tests, so incomplete results are not read as proof that a system is secure.
Threat Models tie attack scenarios to evidence from autonomous tests
Teams can map application components, system relationships, sensitive data and attack paths for existing or in-development systems. Ares converts relevant scenarios into testing objectives. It then records successful attacks, unsuccessful attempts and scenarios that need more testing. Customer-authored scenarios are preserved, and AI-proposed changes to system diagrams require operator review.
In Assail's example, Ares discovers an undocumented API endpoint and Cassandra links it to an existing authorization threat. Ares then runs an approved assessment and updates the model, and Cassandra prepares a retest plan for approval after remediation.
Assail Anywhere extends deployment to air-gapped environments
Through Assail Anywhere, the platform and its AI model can run as SaaS, in a customer's own cloud, on-premises, or in disconnected and air-gapped environments. "Cassandra helps teams decide what to test next," said Alissa Knight, CEO of Assail.