Deploy and manage MCP servers in secure containers
An open-source MCP platform that runs any MCP server in an isolated container with identity, policy, and audit controls, on desktop or Kubernetes.
Why it matters
Simplify the deployment and ongoing management of Model Context Protocol (MCP) servers by packaging them in secure, consistent containers that eliminate configuration drift and security vulnerabilities.
Outcomes
What it gets done
Package MCP servers into isolated Docker containers for consistent deployment
Manage server configurations and secrets securely across environments
Automate the deployment pipeline for multiple MCP server instances
Monitor and maintain containerized MCP servers with standardized tooling
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/mcp-toolhive | bash Overview
ToolHive
An open-source MCP platform with four components - Gateway, Registry Server, Runtime, and Portal - that runs any MCP server in an isolated container with identity, policy, and audit controls. Use it when you need to run MCP servers with real isolation and centralized policy instead of a raw local process, whether on a developer desktop, Kubernetes, or a self-hosted enterprise registry.
What it does
ToolHive runs every MCP server in an isolated container, enforcing identity and access policy per request when configured with an authentication source, and gives platform teams the observability needed to run MCP in production. It's built on four modular components: a Gateway that exposes dedicated endpoints, orchestrates multiple tools into a virtual MCP via a deterministic workflow engine, enforces access policy and SSO (OIDC/OAuth) through your IdP, and filters and customizes tool descriptions to cut token usage; a Registry Server that curates a catalog of trusted MCP servers, integrates with the official MCP registry, and verifies provenance and signs servers; a Runtime that deploys and manages MCP servers locally via Docker or Podman or in a Kubernetes cluster, proxies remote MCP servers, and exports OpenTelemetry and Prometheus telemetry; and a Portal - a cross-platform desktop app plus a browser-based cloud UI - that lets admins curate servers and lets end users install one with a single click. For individual developers, ToolHive wraps every MCP server in an isolated container with a minimal permission file (no local credentials stored) and uses semantic tool search to cut token usage by up to 85%.
When to use - and when NOT to
Use ToolHive when you need to run third-party or internal MCP servers with real isolation, identity enforcement, and audit logging instead of trusting a raw local process - whether that's one developer running a server from the desktop app or CLI, a platform team running MCP on existing Kubernetes infrastructure via the Kubernetes Operator, or an enterprise that needs to self-host its MCP registry and gateway because compliance rules out SaaS MCP providers. It is not itself an MCP server - it is the container, policy, and registry layer that sits in front of whichever MCP servers you actually want to run, and it explicitly targets teams already dealing with "shadow MCP use" that needs centralized policy and audit visibility. The components also ship as separate repos for teams that want to adopt pieces independently - toolhive-registry-server for the catalog, toolhive-studio for the desktop UI, and toolhive-cloud-ui for the browser-based portal - rather than the whole platform at once.
Inputs and outputs
Individual developers connect existing clients (Claude Code, Cursor, GitHub Copilot, and others) to MCP servers with a single click or command via the desktop app or CLI; the Kubernetes Operator instead exposes Custom Resource Definitions for MCP servers, registries, and other ToolHive components, with automated service creation and discovery plus ingress integration. Output on the observability side includes OpenTelemetry traces, Prometheus metrics, and audit logs; on the security side, per-request identity and access enforcement via an IdP (OIDC/OAuth SSO) and secure token exchange. The built-in MCP Inspector and MCP Optimizer let you test and validate a server and tune its token usage before deploying it further.
Integrations
Connects to existing AI clients including Claude Desktop, Claude Code, Cursor, GitHub Copilot, VS Code, and VS Code Server; integrates with the official MCP registry as well as custom MCP server catalogs; runs MCP servers via Docker or Podman locally, or via a dedicated Kubernetes Operator in a cluster; and exports telemetry through OpenTelemetry and Prometheus. Licensed under Apache 2.0.
Who it's for
Individual developers who want a safer, cheaper way to run MCP servers locally; platform engineers running MCP on existing Kubernetes infrastructure who need centralized policy, SSO, and audit logging; and enterprises whose compliance requirements rule out SaaS MCP providers and need to self-host the entire registry-gateway-runtime stack.
Source README
The open source MCP platform trusted by developers and enterprises
[![Release][release-img]][release] [![Build status][ci-img]][ci]
[![Coverage Status][coveralls-img]][coveralls]
[![License: Apache 2.0][license-img]][license]
[![Star on GitHub][stars-img]][stars] [![Discord][discord-img]][discord]
Run any MCP server securely, instantly, anywhere.
ToolHive runs every MCP server in an isolated container, enforces identity and access policy per request when configured with an authentication source, and gives platform teams the observability they need to put MCP in production.
Why ToolHive?
Here are some of the more common use cases for ToolHive:
| Developers. Run MCP servers with more security and more (token) savings | Platform Engineers. Run MCP on your existing Kubernetes infrastructure | Enterprises. Self-host MCP servers and stay in control of your data |
| Connect Claude Code, Cursor, GitHub Copilot, or your preferred client to MCP servers with a single click or command. ToolHive wraps every MCP server in an isolated container with a minimal permission file (no local credentials) and uses semantic tool search to reduce your token usage by up to 85%. |
Put an end to shadow MCP use by your developers, and give your security team the audit logs and configurable identity enforcement they require. ToolHive includes a Kubernetes operator, so you can declare policies, integrate with your IdP and observability stack, emit OTel traces, and more … all with familiar tools and patterns. |
Most MCP solutions are SaaS, but your compliance requirements prohibit sensitive info from being processed by SaaS providers. ToolHive is the exception that allows you to self-host your MCP registry, gateway, etc. You can pilot the entire platform, and when you’re ready to scale, Stacklok’s got the added capabilities and expert team ready! |
| Download ToolHive and get started | Explore the Kubernetes operator in our docs Read more about running MCP on Kubernetes |
Learn more about Stacklok’s platform Compare open source ToolHive and Stacklok Enterprise |
Quick links
- 📥 Downloads
- 📚 Documentation
- 🚀 Quickstart guides:
- 💬 Discord
- 🤝 Contributing
Stacklok Enterprise
Core capabilities
ToolHive architecture: Gateway, Registry Server, Runtime, and Portal
ToolHive is built on a modular architecture to streamline secure MCP server management and integration. Here's how the main components work.
🔌 Gateway
Define dedicated endpoints from which your teams can securely and efficiently access tools.
- Orchestrate multiple tools into a virtual MCP with a deterministic workflow engine
- Define access policies and network endpoints
- Centralize control of security policy, authentication, authorization, auditing, etc.
- Integrate with your IdP for SSO (OIDC/OAuth compatible)
- Customize and filter tools and descriptions to improve performance and reduce token usage
- Connect with local clients like Claude Desktop, Cursor, VS Code, and VS Code Server
📦 Registry Server
Curate a catalog of trusted servers your teams can quickly discover and deploy.
- Integrate with the official MCP registry
- Add custom MCP servers
- Group servers based on role or use case
- Manage your registry with an API-driven interface (or embed in existing workflows for seamless integration and governance)
- Verify provenance and sign servers with built-in security controls
- Preset configurations and permissions for a frictionless user experience
⚙️ Runtime
Deploy, run, and manage MCP servers locally or in a Kubernetes cluster with security guardrails.
- Deploy MCP servers in the cloud via Kubernetes for enterprise scalability
- Run MCP servers locally via Docker or Podman
- Proxy remote MCP servers securely for unified management
- Kubernetes Operator for fleet and resource management
- Leverage OpenTelemetry and Prometheus for monitoring and audit logging
💻 Portal
Simplify MCP adoption for developers and knowledge workers across your enterprise
- Cross-platform desktop app and browser-based cloud UI
- Make it easy for admins to curate MCP servers and tools
- Automate server discovery
- Install MCP servers with a single click
- Compatible with hundreds of AI clients
How it works together
- Admins curate and organize MCP servers in the Registry, configuring access and policies.
- Users discover and request MCP servers from the Portal, and ToolHive orchestrates installation and access.
- Runtime securely deploys and manages MCP servers across local and cloud environments, integrating seamlessly with existing SDLC workflows, exporting analytics, and enforcing fine-grained access control.
- Gateway handles all inbound traffic, secures context and credentials, optimizes tool selection, and applies organizational policies.
Flexible deployment
Desktop experience
Individual developers can get started in minutes with the desktop UI or CLI, then apply the same concepts in enterprise environments.
Key features:
- Run any MCP server from a container image, or build one dynamically from common package managers
- Manage encrypted secrets and control network isolation with simple, local tooling
- Test and validate MCP servers using built-in tools like the official MCP Inspector
- Optimize token usage and tool execution with the MCP Optimizer
Get started with the UI: Quickstart, How-to guides
Get started with the CLI: Quickstart, How-to guides, Command reference
MCP guides: learn how to run common MCP servers with ToolHive
Kubernetes Operator
Teams and organizations manage MCP servers and registries centrally using familiar Kubernetes workflows.
Key features:
- Custom Resource Definitions for MCP servers, registries, and other ToolHive components
- Secure execution with container-based isolation and multi-namespace support
- Automated service creation and discovery, with ingress integration for secure access
- Enterprise-grade security and observability: OIDC/OAuth SSO, secure token exchange, audit logging, OpenTelemetry, and Prometheus metrics
- Hybrid registry server: curate from upstream registries, dynamically register local MCP servers, or proxy trusted remote services
Get started: Quickstart, How-to guides, CRD reference, Example manifests
Hybrid
ToolHive's complete solution for teams and enterprises supports MCP servers across all environments: on developer machines, inside your Kubernetes clusters, or hosted externally by trusted SaaS providers.
End users access approved MCP servers through a secure, browser-based cloud UI. Developers can also connect using the ToolHive CLI or desktop UI for advanced integration and testing workflows.
Enterprise teams can also leverage ToolHive to integrate MCP servers into custom internal tools, agentic workflows, or chat-based interfaces, using the same runtime and access controls.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.