Introducing agentdesktop
Discover how agentdesktop brings visibility, identity, policy, and credential management to AI tools, MCP servers, and agents across your organization’s desktop fleet.
introducing-agentdesktop
Understanding the agentregistry Project and the Problems It Solves
Discover how agentregistry provides a unified, open source catalog for agents, MCP servers, skills, prompts, plugins, and models, with tools to build, publish, discover, and deploy agentic artifacts.
ai agent registry | agentregistry
understanding-the-agentregistry-project-and-the-problems-it-solves
What Is an Agent Sandbox? A Guide to Isolated Execution for AI Agents
An agent sandbox is an isolated runtime where AI agents execute code and tools safely. Learn how sandboxing works and why agentic infrastructure needs it.
what-is-an-agent-sandbox-a-guide-to-isolated-execution-for-ai-agents
Installing kagent and agentgateway from Google Cloud Marketplace
Install kagent and agentgateway from Google Cloud Marketplace on GKE. Deploy AI agents with guided setup, including keyless Gemini access through Vertex AI.
installing-kagent-and-agentgateway-from-google-cloud-marketplace
Installing kagent and agentgateway from Azure Marketplace
Install kagent and agentgateway from Azure Marketplace on AKS. Follow this guide to configure prerequisites, deploy cluster extensions, and verify your installation.
installing-kagent-and-agentgateway-from-azure-marketplace
Installing kagent and agentgateway from AWS Marketplace
Install kagent and agentgateway from AWS Marketplace using EKS add-ons or Helm. Follow this guide to configure prerequisites and deploy both on Kubernetes.
installing-kagent-and-agentgateway-from-aws-marketplace
Agents On Every Cloud
Deploy kagent and agentgateway from AWS, Azure, or Google Cloud Marketplace, then connect them to expose, secure, and observe AI agents on Kubernetes.
agents-on-every-cloud-aws-azure-google-cloud-marketplace-kubernetes
Unit Economics of Agentic AI: Context × Turns × Model
Cost optimization requires more than reporting and budget controls. Controlling the inputs that drive token consumption cuts spend and improves quality at the same time.
unit-economics-of-agentic-ai-context-turns-model
MCP Stateless Spec Changes: The Engineering Details
The largest MCP revision since launch ships July 28: a stateless core, required routing headers, six authorization SEPs, and deprecated Roots, Sampling, and Logging.
mcp-stateless-spec-changes-the-engineering-details
Runaway Tokens Are a Governance Problem
Why unbounded AI spend belongs on the risk register, and how budgets become a real control.
runaway-tokens-are-a-governance-problem
