CAMBRIDGE, Mass., Sept. 03, 2026 (GLOBE NEWSWIRE) -- Solo.io, the leading provider of cloud native connectivity and AI-ready infrastructure, today announced agentdesktop, a new open source project that brings the governance capabilities Solo.io has implemented for agentic infrastructure all the way out to the end user machines where Claude Code, Codex, and other general-purpose agents run.
Agentic AI has introduced a completely new dimension to desktop exploits, increasing both the coverage area and blast radius of incidents. Yet most organizations feel that their existing governance controls via MDM and firewall solutions cover the complete threat surface. A recent study showed 90 percent of executives were confident they know what AI is running in their organization. The reality of the situation is quite different: 52 percent of employees admitted to using AI tools their company never approved (Okta) and 86 percent of enterprises do not enforce access policies for AI identities (Cloud Security Alliance).
The fundamental tradeoff with agentic AI is that the more autonomy and power we give agents, the less predictable and controllable they become. This is exactly why so many agentic projects are able to demonstrate compelling value in the proof-of-concept and pilot phase, but fail to make it to production due to a lack of visibility, accountability, and control. Solo.io calls this the production gap: the distance between the governance organizations have in their infrastructure today and what they need to deploy agentic AI to production. This production gap now extends to the desktop, with agent harnesses connecting to production data, APIs, and systems. Every organization's governance strategy must now extend to this new frontier.
Agentdesktop gives platform and security teams four capabilities from a single deployment:
- Discovery: build a real-time inventory of every harness, model, MCP server, and skill across the fleet, attributed to a device and the associated user. Know exactly what's running and where.
- Policy: declare configuration centrally and agentdesktop applies it to every harness, previews every change, and continuously reconciles drift. Ensuring consistent and uniform security posture across every installation.
- Identity: treat agents as non-human identities with least-privilege access. Devices register, bind to their user through OIDC, and receive short-lived, just-in-time credentials identifying the specific combination of device, user, and agent. Eliminate the blast radius and unpredictable attack vectors of long-lived secrets stored locally.
- Observability: opt-in session and tool-use telemetry, attributed to user and device, from only the events administrators select. Real-time awareness of agent activity.
Users can start on a single desktop in standalone mode to provide local governance across installed agent harnesses, then move to fleet mode, where MDM installs the endpoint daemon and a central controller integrates with the organization's existing identity provider, PKI, and LLM gateway. Because agentdesktop is open source under the Apache 2.0 license, organizations can see exactly what the endpoint daemon collects, audit its identity and credential boundaries, and extend integrations as AI tools evolve. They keep their existing identity provider, PKI, and gateway rather than adopting a vertically bundled platform.
The Solo.io agentic stack provides everything teams need to develop, deploy, and operate AI agents across agentic runtime environments. Powered by core open source projects created and contributed to the community by Solo.io, including:
- agentdesktop is the open source governance layer for AI agents on the desktop, bringing inventory, managed configuration, non-human identity, and observability to the workstations where most enterprise agent activity actually happens.
- kagent, a CNCF Sandbox project, is the Kubernetes-native runtime that makes agents first-class workloads, with the security, observability, and lifecycle controls enterprises need to run programmatic, declarative, and always-on assistant patterns side-by-side.
- agentgateway, an Agentic AI Foundation project, is the high-performance data plane for agentic traffic, built from the ground up in Rust to span the full range of agentic protocols and use cases as a unified inference gateway, LLM gateway, MCP gateway, and A2A gateway.
- agentregistry, contributed to CNCF, is the trusted source for discovering, packaging, and distributing agents and tools across an organization with centralized observability, security, and governance across agent runtimes.
- agentevals captures evaluation and quality signals using OpenTelemetry rather than in-process instrumentation, providing continuous, framework-agnostic visibility into how agents actually behave in production across the entire agentic loop.
To join the community and get started, please visit:
- The project announcement blog for a deeper dive into the project’s capabilities.
- The agentdesktop project site.
- Read the documentation.
- Get involved by contributing via GitHub.
- Join the #agentdesktop discussion in Discord.
About Solo.io
Solo.io is reimagining infrastructure for cloud and AI, uniting secure, seamless cloud connectivity with AI-ready, agentic infrastructure. Trusted by leading enterprises worldwide, Solo.io helps organizations securely connect applications, services, and AI workloads across any environment. From AI infrastructure to API gateways and service mesh, our solutions simplify and unify application networking, enabling teams to accelerate innovation, scale intelligently, and leverage the full potential of modern AI agents. Learn more at www.solo.io.
Media Contact
Solo.io Press Team media@solo.io
