August 5, 2025
5 min read
Mike Vizard
Cycode’s new AI agent ranks vulnerabilities by exploitability, helping DevSecOps teams prioritize remediation faster and more effectively.
Cycode has introduced an AI agent to its application security posture management (ASPM) platform, designed specifically to assess how exploitable a vulnerability found in an application truly is.
In addition to this AI Exploitability Agent, Cycode has released an AI Security Return on Investment (ROI) Calculator that evaluates the impact of AI on various DevSecOps use cases. Devin Maguire, senior product marketing manager at Cycode, explained that this AI agent simplifies prioritization of remediation efforts by helping DevSecOps teams focus on vulnerabilities that pose the greatest risk to their organizations.
This innovation comes at a critical time when AI coding tools are generating more vulnerabilities than ever. Cycode estimates that one security flaw is created for every 10,000 lines of code written. Furthermore, approximately 40% of AI-generated applications contain some form of vulnerability. More concerning is that cybercriminals are leveraging AI capabilities to discover and reverse engineer these vulnerabilities, increasing the threat landscape.
The Cycode AI Exploitability Agent is part of a broader suite of AI Security Teammates introduced earlier this year. These include:
- Change Impact Analysis Agent: Monitors code changes across pull requests to identify significant risk-altering changes.
- Fix & Remediation Agent: Analyzes root causes of issues and suggests code fixes. These AI agents integrate with Cycode’s Risk Intelligence Graph (RIG), which aggregates data from code repositories, workflows, secrets, dependencies, and cloud infrastructure assets. Support for the Model Context Protocol (MCP), an emerging AI integration standard developed by Anthropic, enables these agents to access and share data seamlessly. This allows for correlated scans and consolidated alerts, improving efficiency. The ultimate goal is not only to speed up vulnerability identification and remediation but also to enhance collaboration between application development and cybersecurity teams. Legacy application security tools often flag vulnerabilities in code that is inaccessible or never loaded into memory. In contrast, AI agents provide richer context by analyzing both code and runtime environments, enabling more accurate risk assessment. A recent Futurum Group survey highlights that investments in ASPM platforms, DevSecOps automation, and orchestration are top priorities for organizations. Interestingly, security budgets are increasingly shared, with only 21% of respondents indicating that security teams solely fund these initiatives. Half of the respondents reported that application development teams now share responsibility for application security. As global regulations tighten, securing software supply chains will become even more essential. The challenge lies not in whether applications will become more secure, but in how quickly and at what cost.
- AI Agents Capabilities, Risks, and Growing Role
- AI-Driven Crypto Scams Surge 456%, Experts Warn No One Is Safe
- AI Crypto Convergence: 2025 Market Gains
Source: Cycode Delivers AI Agent to Assess How Exploitable Vulnerabilities Are by Mike Vizard, August 5, 2025