Praxen in the news

The New Insider Threat: Why AI Agents Need Governance Before They Need Access — Exabeam's Mazen Dohaji on verifying agent behavior with Praxen before granting enterprise access

Exabeam launches Praxen, an open-source tool to verify AI agent behavior

Praxen: Open-source AI agent behavior verification

Exabeam releases Praxen, an open-source tool for verifying AI agent behaviour before deployment

Exabeam launches open-source Praxen to verify AI agents

Exabeam launches open-source Praxen to verify AI agents

Exabeam releases open-source Praxen tool for AI agent behaviour verification

Exabeam unveils open-source Praxen to establish Agent Behavior Verification for enterprise AI workers

Exabeam launches open-source Praxen for Agent Behavior Verification

Exabeam launches Praxen for AI agent security

Exabeam launches open-source Praxen for AI agent verification

Praxen featured among the hottest open-source cybersecurity tools of June 2026

Exabeam releases open-source Praxen to verify AI agent behaviour and security

Praxen: Open-Source AI Agent Behavior Verification

Praxen featured among 20 open-source cybersecurity tools to keep your team ready

Praxen: Open-Source AI Agent Behavior Verification Explained

Exabeam Introduces Open Source Tool to Bring Agent Behavior Verification to AI Agents and Digital Workers

4 Open-Source Tools to Secure AI Agents — Praxen named the behavior-verification layer of an emerging "AI Agent Security Stack," checking declared policy against code, configuration, logs, and governance docs to catch capability drift

4 Free Tools to Secure AI Agents — Praxen featured in the cybersecurity keynote speaker's weekly practical-resources series (video reel)

When a Security Scan Earns Its Keep — a developer ran Praxen against a security-sensitive agent codebase: a data-exposure gap in diagnostic exports, unsafe tool descriptions, and unescaped third-party text in a privileged instruction channel, all found and fixed the same day

What Praxen Found on FinBot and Where SAILORS Fits — a walkthrough of Praxen's FinBot findings, showing how design-time (SAILORS) and implementation-time (Praxen) checks catch the same gaps

Frontier AI is Changing Cyber Risk: Agent Behavior Needs a New Approach

AI agents need a role and a scope too — verify what they actually do, not just what they're documented to do

What Happens When You Actually Test Whether Your Clinical AI Does What It Claims

Behavioral Verification Is Not Optional for AI Agents in Regulated Industries

The Scanner, the SOC Agent, and the Prompt That Lied — a Praxen maturity scan (1.85 → 2.75) shows why deterministic guardrails beat prompt-only enforcement

When OWASP LLM Risks Meet Agentic Risks — the LLM and Agentic Top 10s are complementary layers, shown through Praxen 1.1 testing where Excessive Agency and Unexpected Code Execution rise to the top

AI agent behavior verification, achieved in open source — this could change the future of AI (Japanese-language thread)

Running agents without behavior observation is a gamble — a Verify-Observe-Analyze-Improve playbook built on Praxen remit verification and Observra telemetry standardization (Japanese-language thread)

Exabeam launched Praxen — agent verification as a Claude Code plugin, built by the OWASP Gen AI Security chair: checks permissions, boundaries, and controls pre-deployment, every release

Praxen brings Agent Behavior Verification to open source — shifting AI agent security from post-deployment detection to pre-deployment checks on permissions, approved systems, and business purpose