Ox Security has launched an AI-powered tool called Agent Ox that not only identifies software vulnerabilities but also generates custom code to fix them.
The platform integrates with existing security tools and highlights issues for developers. However, since developers often juggle multiple requests, prioritizing bug fixes can be difficult. Previously, Ox provided general suggestions like “avoid passing user input directly to your database,” but these lacked specific implementation steps.
Agent Ox now takes it further by analyzing the organization’s code and generating tailored fixes that match the developer’s style and codebase. Developers can review the code and approve it with a single click, pushing the changes to repositories like GitHub and into production during the next CI/CD cycle.
The system works in three stages: identifying vulnerabilities, filtering out false positives, and generating secure, context-aware fixes. Agent Ox uses a team of AI agents to evaluate issues from various perspectives, such as understanding business logic, data sensitivity, and architecture.
According to Ox CEO Neatsun Ziv, most AI tools offer generic solutions, but Agent Ox provides intelligent, context-specific fixes that build developer trust. It enables a focused form of AI-driven coding where each fix is small, manageable, and directly integrated into the developer’s workflow.
Ziv believes that while broader AI-based coding systems are still evolving, tools like Agent Ox mark a practical step toward intelligent, automated code maintenance.
Found this article interesting? Follow us on X(Twitter) ,Threads and FaceBook to read more exclusive content we post.
