Blog

Updates, insights, and MCP know-how from the team at Stacklok

October 28, 2025

Cut token waste from AI workflows with the ToolHive MCP Optimizer

MCP Optimizer uses smart tool selection to reduce token usage by 85%. Give your context window a break by loading relevant tools only when they’re needed.

October 14, 2025

Test and tune MCP servers instantly in the ToolHive Playground

Simplify your AI agent development with instant validation, detailed logging, and conversational server management. Supports multiple AI model providers.

October 06, 2025

Seamless remote MCP server support with ToolHive

Connect to remote MCP servers with native OAuth/OIDC support and dynamic client registration. Unified commands work for both local and remote servers.

September 30, 2025

From black box to observable: Deploy ToolHive with OTel + Prometheus

Hands-on tutorial for deploying ToolHive with OpenTelemetry, Prometheus, and Grafana in Kubernetes. Transform black-box MCP servers into observable services.

September 25, 2025

Bridging the observability gap in MCP servers with ToolHive

ToolHive collects usage statistics from MCP servers and feeds data into OTel and Prometheus. Get full visibility with request metrics and distributed traces.

September 22, 2025

The next big observability gap for Kubernetes is MCP servers

MCP servers create blind spots in Kubernetes monitoring. They lack metrics endpoints, don’t log structured data, and can’t be traced by Prometheus or OpenTelemetry.