ClearRoute launches open-source agentic AI control plane for software delivery

ClearRoute, a global platform engineering consultancy, has launched runkora, an open-source, cloud-agnostic agentic AI control plane designed to help organisations govern AI agents throughout the software development and delivery lifecycle.

Runkora was built by ClearRoute's Australian engineering team and is now available in alpha.

The company said the platform addresses compliance and governance challenges that arise as organisations deploy growing numbers of AI agents in software engineering and DevOps workflows.

ClearRoute said runkora operates as a common control plane - rather than a per-agent gateway - enforcing controls on agents themselves and confining them to authorised environments.

The platform covers eight governance areas: identity and delegation, context and discovery, policy and authorisation, secure credentials and access, isolated execution, model and inference access, enterprise connectivity, and observability and audit.

Runkora uses ClearRoute’s platform engineering approach to govern AI agents throughout the end-to-end journey of getting software from initial idea into production and, ultimately, into customers’ hands. This journey is called the Route to Live and includes the processes, tools, and cultural practices that enable or impede delivery excellence.

"We've already seen how AI can speed up writing code, but that's only 20% of the software development lifecycle. The bottleneck occurs with what comes next, because the remaining 80% requires testing, security, compliance, governance and release orchestration before any software can safely go into production," said James Jarvis, chief executive at ClearRoute.

Jarvis said ClearRoute's Route to Live research shows some organisations are still waiting four to eight weeks for change approvals, meaning faster code generation does not automatically translate into faster time to market.

The credential management burden at scale is a central problem runkora aims to solve.

Justin Wilkin, global head of engineering and AI at ClearRoute, said that with just ten agents deployed, organisations rotate 20 to 30 secrets per month - a figure that becomes unmanageable as agent counts climb.

Gartner predicts the average global Fortune 500 enterprise will have more than 150,000 AI agents in use by 2028, the company noted, while only 13% of organisations believe they have adequate AI agent governance in place.

"runkora solves this by centrally controlling agent identities, credentials, permissions and actions, confining agents to authorised environments while giving humans visibility of any attempted activity outside those boundaries," Wilkin said.

Wilkin also distinguished runkora from what he described as agentic AI gateways - tools that assume agents will follow prescribed boundaries - arguing that as AI becomes more autonomous, enforceable controls built on zero-trust principles are required instead.

“This is where platform engineering becomes critical, making secure, compliant ways of working the easiest and most consistent path for every agent, rather than expecting individual teams to build and maintain controls themselves," he said.