Cycode has unveiled its Agentic Development Security Platform to secure the shift from human-centric to AI-driven software development. Purpose-built for the Agentic Development Lifecycle (ADLC), the platform introduces Cycode Maestro orchestration alongside new AI visibility, governance, and real-time guardrails. As enterprises adopt agents to write and deploy code, the traditional security model has broken, creating a triple-threat that increases software risk exponentially: development is faster, the attack surface is wider, and adversaries use the same AI tools as engineering teams.
Cycode launched the Agentic Development Security Platform, purpose-built for the Agentic Development Lifecycle (ADLC).
The platform introduces Cycode Maestro orchestration for multi-agent security workflows with minimal human intervention.
New capabilities include AI Governance, Agentic Coding Guardrails, AI Risk Detection, and ADLC Supply Chain Security.
Agentic Coding Guardrails monitor AI agents, prompts, and code outputs in real time to block risky exposure.
AI Risk Detection enforces policies aligned with OWASP LLM Top 10 standards.
Cycode is showcasing the platform at RSA Conference 2026.
“Agentic development demands agentic security,” said Lior Levy, CEO and co-founder of Cycode. “We are moving beyond passive scanning toward proactive, self-protecting security. Our customers believe in what we’re doing, and that’s why this announcement follows another record-breaking quarter. By extending visibility and governance over the AI supply chain, integrating guardrails into agentic development workflows, and orchestrating specialized security agents via Maestro, we are equipping customers to secure their competitive advantage in the AI era.”
As enterprises adopt agents to write and deploy code, the traditional security model has broken. The shift has created a triple-threat that increases software risk exponentially: development is faster, the attack surface is wider, and adversaries use the same AI tools as your engineering team. Security built for human-paced development cannot protect an AI-driven software factory.
The platform’s latest innovations are designed to automate security and expand coverage across the modern AI-native software factory:
Agentic Security (Cycode Maestro): Orchestrates multi-agent security workflows to identify, triage, and remediate exploitable vulnerabilities with minimal human intervention.
AI Governance and ADLC Supply Chain Security: Discovers Shadow AI and provides policy-driven control over AI models, infrastructure, and tools.
Agentic Coding Guardrails: Monitors AI agents, prompts, and code outputs in real-time to block risky exposure and prevent the introduction of vulnerable AI-generated code.
AI Risk Detection: Detects AI-specific risks and enforces policies to ensure adherence to standards like the OWASP LLM Top 10.
“Traditional security tools are blind to the risks of the AI era, like the use of untrusted models, secrets leaked via prompts, or the detection of AI-specific vulnerabilities,” said Dor Atias, Chief Product Officer at Cycode. “Cycode is closing this gap by expanding our platform scope and arming security teams with their own set of agents to secure modern agentic development.”
About Cycode
Cycode’s Agentic Development Security Platform enables enterprises to secure development from prompt to cloud. Its self-protecting ADLC vision brings security, developers, and agents together with actionable context to prevent, prioritize, and fix the risks that matter.
Powered by its convergence of AST, SSCS, and ASPM with deterministic & AI-scanning engines, third-party integrations, and the Context Intelligence Graph, Cycode delivers correlated insight across the entire Software Factory. With Maestro, Cycode manages and orchestrates agents to allow security teams to keep up with the 10X attacker, while built-in AI Governance and guardrails ensure the safe, compliant use of AI across the ADLC so developers can build.