ToolHive - making MCP servers easy and secure

Name: ToolHive - making MCP servers easy and secure
Author: stacklok

@stacklok

代码仓库主页

关于 ToolHive - making MCP servers easy and secure

ToolHive is an enterprise-grade platform for running and managing Model Context Protocol (MCP) servers.

#kubernetes #golang #security #ai #mcp #ai-security #aicodeassistant #mcp-servers #mcp-tools #mcp-security

基本信息

分类

其他

许可证

Apache-2.0

运行时

传输方式

stdio

发布者

stacklok

配置

暂无标准配置

该服务器的 README 中没有可解析的 MCP 配置块，请前往代码仓库查看安装说明。

代码仓库

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is ToolHive?

ToolHive is an open source MCP (Model Context Protocol) platform that runs any 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 needed for production. It is designed for developers, platform engineers, and enterprises who need secure, scalable, and self-hosted MCP server management.

How to use ToolHive?

Install ToolHive via the desktop app, CLI, or Kubernetes operator. Follow the quickstart guides to connect clients like Claude Code, Cursor, or GitHub Copilot with a single click or command. Administrators curate servers in the Registry, users discover them through the Portal, and the Runtime deploys them locally or in Kubernetes.

Key features of ToolHive

Runs every MCP server in an isolated container
Enforces identity and access policy per request
Reduces token usage by up to 85% with semantic tool search
Supports local (Docker/Podman), Kubernetes, and hybrid deployments
Integrates with IdP for SSO (OIDC/OAuth compatible)
Provides OpenTelemetry traces and Prometheus metrics for observability

Use cases of ToolHive

Developers run MCP servers with more security and token savings
Platform engineers run MCP on existing Kubernetes infrastructure
Enterprises self-host MCP servers and stay in control of their data

FAQ from ToolHive

How does ToolHive ensure security?

ToolHive isolates every MCP server in a container with minimal permissions, enforces identity and access policy per request, and provides audit logging and fine-grained access control.

What deployment options does ToolHive support?

ToolHive can run MCP servers locally via Docker or Podman, in a Kubernetes cluster via its operator, or in a hybrid setup combining local, cloud, and externally hosted servers.

Can I self-host ToolHive?

Yes. ToolHive is fully open source and designed for self-hosting. Enterprise teams can run their own MCP registry, gateway, and runtime while staying in control of sensitive data.

Which AI clients are compatible with ToolHive?

ToolHive works with Claude Desktop, Cursor, VS Code, VS Code Server, GitHub Copilot, and hundreds of other AI clients through its Gateway and Portal interfaces.