LAS VEGAS, Aug. 04, 2026 (GLOBE NEWSWIRE) —
As AI-assisted attackers compress exploitation timelines to hours, Mallory turns live adversary intelligence into prioritized, policy-governed action across the tools security teams already run
Mallory, the AI-native Threat and Exposure Management platform, today introduced a unified context and intelligence layer for security teams. The architecture has three parts: a context graph that correlates attack surface, threat, and vulnerability data; an intelligent reasoning layer that determines what matters and why; and a policy and governance layer that routes prioritized work into action under each customer’s own rules. One foundation supports exposure investigation, threat hunting, supply chain risk management, and vulnerability prioritization, rather than locking teams into a single fixed workflow.
“Our vision is an intelligent, agentic platform constantly evaluating whether a threat is actually a risk in your environment, so your team gets a faster, sharper response and spends its time on the risk that matters. The intelligence to hunt threats, prioritize exposures, and build detections all starts from the same data, so it shouldn’t take three tools to make it work. And when a working exploit lands in hours, a queue ranked by a severity score that doesn’t know who’s attacking you is wasting your best analysts on work that should be automated,” said Jonathan Cran, founder and CEO of Mallory.
The context graph, reasoning layer, and policy layer are fully separable, so Mallory fits any security team’s stack, not just one built around it. Teams that just want a contextual data source can plug Mallory’s threat and context data straight into the workflows they already run. Teams ready to go further can adopt Mallory’s agentic harness out of the box, handing routine exposure remediation to agents at scale, all within the policy guardrails they define and control.
Mallory’s consumption model is a deliberate break from usage-based AI pricing. It meters on coverage and outcomes rather than token usage, and supports bring-your-own-key (BYOK), so teams run on their own model infrastructure or Mallory’s. Teams pay for software and the outcomes it produces, not for how much a workload happens to consume.
AI-assisted attackers have collapsed the cost and time of finding exploitable flaws, and security teams are drowning in intel they cannot act on fast enough. Point tools force teams to choose between prioritizing exposures, hunting for threats, or building detections, when the real problem is upstream: knowing what to look for, where to look for it, and being fast and cost-effective enough to act on it. Mallory’s architecture is built to close that gap once, at the layer underneath all three problems, rather than solving each one separately.
- The context graph pulls in attack surface and security configuration information, then unifies it with external threat intel and vulnerability information using the same underlying pipelines. Every new CVE or adversary technique is correlated against an organization’s actual exposure within minutes, not days.
- The intelligent reasoning layer sits on top, determining whether a given signal reaches the environment, where it lands, and how much it matters given real adversary activity, not a static severity score alone.
- The policy and governance layer sits above that, letting teams set exactly how much autonomy the reasoning agents get: auditing code repositories against current adversary techniques, watching supply chain dependencies, evaluating CI/CD configurations, or routing a prioritized case straight into the ticketing tool the team already uses.
Mallory Founder, Jonathan Cran, writes about its origin in the blog: Adversary Timelines Have Collapsed: Defenders Must Rethink Proactive Security with Agents
About Mallory:
Mallory unifies threat intelligence, exposure context, and response into one architecture for security teams. It monitors adversary activity, contextualizes it against an organization’s attack surface, and converts fragmented telemetry into prioritized, evidence-based action. Four components work together: threat intelligence, a unified context graph, an agentic harness that investigates and writes cases, and a policy and governance layer that keeps agent work scoped, auditable, and controlled. Security teams run Mallory on their own models, keys, and infrastructure. Founded in 2024 and headquartered in Austin, Texas, Mallory is backed by Decibel Partners, LiveOak Ventures, and Aviso Ventures. Users can learn more at mallory.ai.
Contact
Head of Marketing
Alexa Rzasa
Mallory
[email protected]
A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/bee8c311-bdee-4ebd-a9c3-634201c13c70
