Terra Security Announces Internal Network Pentesting, Becoming the First Agentic Offensive Security Platform Covering All Major Attack Vectors

Terra Platform™ tests how attackers actually operate across the web, AI, external networks, and now internal infrastructure: a single platform which offers a continuous closed loop from true real-world risk through remediation

NEW YORK, July 21, 2026 /PRNewswire/ -- Terra Security, the leading AI-native Offensive Security platform, today announced that its continuous, agentic internal network pentesting in Terra Platform™ is in preview with select design partners. The capability extends Terra Platform's coverage from web and internal applications, external networks, and AI systems to the most under-tested and vital attack surface in the enterprise – the internal infrastructure. This makes Terra the first agentic Offensive Security provider to run continuous Offensive Security across all four of these business-critical attack surfaces.

Breaches begin in one area of a business and spread from there. The biggest challenge for organizations is tracking and nullifying an exploitable risk before damage occurs. With the rapid adoption of agentic AI, new attack surfaces are constantly emerging, while AI has empowered threat actors to accelerate the time to exploit known vulnerabilities, reducing the window to defend against attacks to mere minutes. Security leaders now need more than just visibility, they need validation of what's truly exploitable. Terra platform provides unparalleled verification and validation into how an attacker moves through an enterprise, from initial access through external networks, web apps, AI systems, and now into the internal infrastructure - the domain where malicious actions cause the most destruction.

"Attackers do not stop at the perimeter, and neither should Offensive Security," said Shahar Peled, Co-Founder & CEO at Terra Security. "For years, the internal network layer has been the least tested in Offensive Security and least understood, invariably leading to breaches and escalation. Expanding Terra Platform to internal infrastructure means that, for the first time, enterprises can continuously and fully answer the most critical question in cybersecurity: if adversaries attack me today, what can they achieve? Not a single vector at a time, or a partial view, but true real-world adversarial emulation with cross attack surface chaining. This is what helps organizations adopt AI safely and fight against AI adversaries confidently. From the beginning, we set the standard for safe, scalable, and accurate continuous offensive security, autonomous by default with a human-ON-the-loop by design."

Traditional internal network testing or scripted, deterministic solutions are not fit for the speed and scale of today's attacks. It often requires Offensive Security professionals to be physically on site or persistent VPN access, complex scoping negotiations, and a test window that only yields a point-in-time snapshot of a continuously changing environment. Moreover, infrastructure-as-code, CI/CD-connected internal services, and AI-powered tooling change internal network topology far faster than annual assessments can track. Terra platform reinvents how internal networks are tested, without impeding regular business operations.

Terra Platform powers continuous, autonomous validation through hundreds of specialized AI agents working in parallel, paired with a Human-on-the-Loop where necessary for sensitive actions in production environments. Agents test and trigger the moment either code or infrastructure changes, chaining findings across attack surfaces the way a real attacker would, exposing what may begin as low-severity issues, but which can lead to a single, prioritized path to compromise. Internal network findings will flow through Terra's existing full triage lifecycle - from exploit to remediation guidance to retest - the same closed-loop process Terra already applies across its other attack surfaces.

Security researchers and pentesters engage with and oversee Terra's AI agents through the Terra Offensive Research Collaboration Hub (TORCH). This hub serves as the controlled gateway through which pentesters and red teamers can gain visibility, investigative capabilities, and structured control before, during, and after testing. It is the space where human judgment and expertise moderates and adjusts an agent's actions as required.

Internal Network Infrastructure capabilities are available on the Terra Platform in a limited preview. Enterprises interested in early access can join the waitlist here.

About Terra Security

Terra Security provides an AI-native Offensive Security platform aligned to code changes and evolving attack surfaces, combining a swarm of trained AI agents for speed and scale with human supervision for safety and control. Fortune 500 organizations trust the Terra Platform™ to find, validate, and help remediate exploitable vulnerabilities across the web, AI systems, internal apps, APIs, mobile, networks, and the cloud. For more information, visit terra.security.

Media contact
[email protected]

SOURCE Terra Security