Harness has introduced Agent DLC, extending software engineering best practices to AI agents by governing the entire AI agent development lifecycle from creation and evaluation through deployment, runtime operations, and continuous improvement. The platform addresses a growing challenge, as enterprises quickly adopt AI agents without the visibility, security, and governance required to manage them at scale.
While software development evolved through established software development lifecycle (SDLC) practices, AI agents have largely been built without comparable operational controls. Agent DLC applies familiar engineering disciplines to AI, providing structured workflows, policy enforcement, evaluation, deployment approvals, runtime monitoring, and auditability throughout an agent’s lifecycle.
The platform introduces capabilities across every stage of the lifecycle, including agent creation, automated evaluation, secure deployment, runtime governance, and continuous optimization. New features include but aren’t limited to Agent Discovery and Posture Management, which automatically identifies and maps AI agents across an organization, and an AI Firewall that protects against prompt injection, tool misuse, and data exfiltration while maintaining a complete audit trail.
Built on the Harness Software Delivery Platform and its Software Delivery Knowledge Graph, Agent DLC extends the same governance, policies, approvals, and evidence already used for software delivery to AI agents. This enables organizations to apply consistent controls from development through production without introducing disconnected security or management tools.
Agent DLC builds on the company’s previously announced Autonomous Worker Agents, expanding governance beyond execution to cover the full lifecycle of enterprise AI agents as organizations increasingly rely on agentic AI in production environments. The new capabilities are rolling out to Harness customers now.
Click Here For The Original Source.
