Radware Adds Governance Reporting and Claude Code Protection to Agentic AI Security Platform

Radware announced an expansion of its Agentic AI Protection platform with new governance reporting, wider AI-agent visibility, and security coverage for developer-hosted agents including Anthropic’s Claude Code.

The company described the update in a July 7 GlobeNewswire release, saying the enhancements are intended to help organizations monitor AI-agent activity, support compliance efforts, and reduce risk as agentic AI tools are adopted in enterprise and developer environments. Help Net Security, SiliconANGLE, and StreetInsider also reported the same product expansion.

What Radware added

According to Radware’s announcement, the updated Agentic AI Protection offering now includes AI governance reporting designed to support compliance and oversight programs. Help Net Security reported that the reporting is aligned with frameworks and regulations including ISO 42001, the EU AI Act, and the NIST AI Risk Management Framework.

The new reporting capability is positioned as “audit-ready,” according to Help Net Security, meaning it is intended to help security and governance teams document AI-agent activity and risk controls for internal reviews or external compliance processes.

Radware also said it expanded visibility across the AI-agent ecosystem. SiliconANGLE reported that the update gives organizations deeper insight into agent activity, while StreetInsider described the enhancement as expanded agent monitoring. The shared emphasis across the reports is that Radware is trying to help security teams see how agents are behaving across enterprise systems, rather than treating AI activity as a black box.

Claude Code and developer endpoint protection

A notable part of the update is new protection for developer-hosted AI agents, including Anthropic’s Claude Code. Radware’s GlobeNewswire release specifically cited Claude Code protection, and SiliconANGLE reported that the coverage applies to agents running on developer machines.

That focus reflects a practical security concern: AI coding assistants and agentic developer tools may operate locally or within development environments where they can interact with code, credentials, repositories, and command-line workflows. Radware’s update appears aimed at giving organizations a way to apply monitoring and controls to those developer-side AI activities.

Help Net Security reported that Radware’s new protection applies to Claude Code on developer endpoints. StreetInsider likewise reported coverage for developer-hosted AI agents including Anthropic Claude Code.

Why governance is part of the security pitch

Radware is framing the update around both security and governance. The company’s release emphasized compliance reporting alongside technical protections, while Help Net Security connected the feature set to major AI governance frameworks and regulations.

That pairing is increasingly common in enterprise AI security products. Organizations adopting AI agents need controls that address not only misuse or compromise, but also documentation, oversight, and accountability. Based on the sources, Radware’s update is meant to serve both needs: monitor agent behavior and generate evidence that can support governance obligations.

The reports do not provide independent testing results or customer deployment data for the new capabilities. They do, however, consistently describe the same core additions: compliance-focused reporting, expanded visibility into AI-agent activity, and protection for developer-hosted agents such as Claude Code.

Availability and market context

Radware did not use the cited reports to disclose detailed pricing or technical benchmarks for the expanded Agentic AI Protection features. The company positioned the update as part of its broader application and API security portfolio, according to its GlobeNewswire announcement.

For security teams, the most relevant takeaway is that Radware is extending agentic AI security beyond centrally managed applications and into developer workflows, while adding reporting intended to map AI-agent controls to governance frameworks. Whether that reduces operational risk will depend on implementation details, integration with existing security tooling, and the scope of agent activity an organization can actually observe.