Stacklok is an enterprise MCP (Model Context Protocol) platform that enables organizations to securely run and manage AI agents connected to enterprise systems and data. Built by the co-creators of Kubernetes, it extends Kubernetes clusters into a first-class MCP runtime, providing oversight, control, compliance, and identity integration. The platform includes a Registry for curating trusted MCP servers, a Runtime for deploying servers with security guardrails, a Gateway for a single endpoint to access tools, and a Portal for admin control and user access. Stacklok helps platform engineering, AI infrastructure, and security teams deploy AI agents in production with standardized connectors, policy enforcement, and visibility. It is built on the open-source ToolHive project (Apache 2.0) and integrates with existing cloud native tools and observability stacks.
Key Benefits
- Built by Kubernetes co-creators
- Open-source core under Apache 2.0
- Native OTel instrumentation for observability
- Policy-as-code with GitOps workflows