As organizations continue to adopt autonomous AI systems, the importance of AI Agent Behavior Verification is increasing. Exabeam announced Agent Behavior Verification (ABV) and released Praxen, an open source reference implementation that helps companies test the security of AI agents prior to deploying them in production.
Exabeam, the leader in Behavior Intelligence for the agentic enterprise, today announced Agent Behavior Verification (ABV), a new security discipline to help organizations verify that AI agents are properly configured, authorized and governed before entering operational environments.
The company also announced Praxen, an open source reference implementation of ABV. The solution allows organizations to evaluate AI agents on permissions, tools, integrations and security controls. Praxen also helps teams validate that agent capabilities match their expected responsibilities.
As AI agents transition from digital assistants to operational systems, enterprises are facing a new set of cybersecurity challenges. These agents can interface with business systems, leverage connected tools, execute workflows and make decisions with increasing autonomy.
However, these emerging risks are not fully addressed by traditional security approaches. Current solutions, including vulnerability assessments and red team exercises, enable organizations to observe and test agent activity during runtime. But companies need better ways of checking if AI agents are safe before they are unleashed.
Agent Behavior Verification addresses this need by evaluating AI agents as complete operational systems. ABV does not just look at software vulnerabilities or individual snippets of code, but rather assesses if an agent’s permissions, implementation and security controls are consistent with its intended purpose.
“Organizations are rapidly moving from AI experimentation to operational deployment,” said Steve Wilson, Chief AI Officer at Exabeam and Founder and Co-Chair of the OWASP Gen AI Security Project. “As agents become digital workers, security teams need more than runtime visibility. They need confidence that agents have the right permissions, the right controls, and the right boundaries before they enter production. Agent Behavior Verification helps answer a fundamental question: will this agent do its job, and only its job?”
Praxen Strengthens AI Agent Security Governance
Praxen operationalizes Agent Behavior Verification through an ABV remit, which works as a policy contract. This contract defines what an AI agent can perform, which resources it can access, and the operational limits it must follow.
Furthermore, Praxen helps developers and security teams evaluate an agent’s implementation, tools, configurations, memory, integrations, and operating environment. The platform identifies differences between expected and actual agent behavior.
As a result, organizations receive actionable recommendations, security findings, and maturity scoring. These insights help teams improve AI security posture before deploying autonomous systems.
“Traditional security tools help identify vulnerabilities in software,” continued Wilson. “Praxen evaluates something different: whether an agent’s capabilities, permissions, tools, and controls align with the role it was authorized to perform. This addresses one of the most critical risks introduced by highly autonomous agents and establishes a stronger foundation for ongoing governance throughout the agent lifecycle.”
Agent Behavior Verification supports Exabeam’s broader agent security strategy. It complements ABA, which helps organizations detect unusual or risky agent behavior after deployment.
Built as an agentic coding agent skill, Praxen is released under the Apache 2.0 license. Moreover, the open-source framework allows developers, researchers, and security professionals to examine, extend, and apply ABV practices within their environments.
“Most security tools tell you what’s vulnerable. Praxen asked a different question entirely: Does this agent’s actual behavior match the governance or work remit it was built to enforce?” said Sherri Douville, CEO of Medigram. “The code-level remediation path it produced didn’t give us a risk report to file away. It gave us a precise engineering roadmap we could act on immediately. In enterprise AI deployment, the gap between what an agent is authorized to do and what it is actually capable of doing is where operational risk lives.”
Building Trust in Autonomous AI Systems
Exabeam is releasing Praxen as an open-source project to encourage wider adoption of Agent Behavior Verification practices. The company aims to support transparency and collaboration across the AI security ecosystem.
Currently, the industry continues developing standards for AI agent governance, monitoring, and verification. Therefore, open-source initiatives like Praxen provide researchers and security teams with opportunities to improve AI governance frameworks.
Through Praxen, organizations can better understand AI agent capabilities, strengthen security controls, and reduce operational risks. Additionally, the framework supports responsible deployment of autonomous AI systems across enterprise environments.
Exabeam’s latest initiative highlights the growing need for AI security solutions that provide visibility, accountability, and trust. As enterprises expand their use of digital workers, Agent Behavior Verification can become an essential foundation for secure AI adoption.
To explore how Security Operations Centers (SOC) play a crucial role in defending against modern cyber threats, read our latest SOC News.
Source: Businesswire
