Skip to main content

Agents arrive in the Playground

This week, ToolHive Desktop UI v0.35.0 makes Agents in the Playground available to everyone, with two built-in agents and the option to author your own. The Desktop UI also retires the localhost TCP port it used to talk to ToolHive in favor of a per-user socket, and the Playground picks up message editing, queueing, and per-message cost.

Registry claims enforcement and horizontal scaling

This week brings Registry Server v1.2.0 with stronger access control over published entries and a clean break from the legacy format. Virtual MCP Server (vMCP) and MCP servers gain horizontal scaling support with Redis-backed session routing to keep sessions consistent across replicas.

Reusable agent skills across CLI and Registry

ToolHive now supports agent skills - reusable bundles of instructions and configuration that teach AI agents how to perform specific tasks. MCP servers give agents the raw tools they can call; skills give them the knowledge of when, why, and how to use those tools effectively. With ToolHive, teams can create skills, publish them to a registry, and install them across supported AI clients from a single CLI.

Token optimization goes Kubernetes-native

MCP Optimizer capabilities are now embedded directly in vMCP, bringing the same token reduction and improved tool selection per request from desktop to Kubernetes-native deployments. Instead of every developer configuring a local optimizer instance, platform teams deploy it once and every connected client benefits automatically.

Enterprise identity and production resilience

This week brings enterprise readiness and resilience improvements across ToolHive. The new embedded authorization server lets users authenticate to MCP servers through their company identity provider, no locally stored credentials required. vMCP adds circuit breakers that prevent cascading failures when backends degrade, and the Registry Server gains cluster-wide namespace scanning for multi-tenant Kubernetes deployments.

Registry observability and zero-downtime operations

This week delivers capabilities that production teams need: vMCP now auto-discovers backend servers without restarts and reports operational health in CRD status fields, the Registry gains OpenTelemetry support and tool definition APIs, and both the CLI and Desktop UI add custom header configuration for remote MCP connections. We've also added custom CA support for organizations running private identity providers.