In one sentence
An open-source, self-hosted Gateway for a personal AI assistant that connects models, tools, sessions and messaging channels.
OpenClaw is an open-source project that runs a Gateway on a machine you control, such as a Mac, a Linux server or a cloud VPS. The Gateway connects AI model providers, tools, sessions, messaging channels such as Slack, Discord, Telegram or WhatsApp, a Control UI and optional device nodes. It can be extended with skills, plugins and MCP servers, and it can run scheduled jobs and webhooks. Its documentation lives at docs.openclaw.ai and its code on GitHub.
Who it is for
OpenClaw is designed around a trusted operator: one person or a small trusted group using their own assistant. Its documentation is clear that one Gateway is not a security boundary between users who do not trust each other. Separate Gateways, operating-system users or hosts are the safer design when trust boundaries differ.
What setup involves
Installing it is quick. The real decisions come after: where the Gateway runs, who can reach it, which channels and senders are allowed, which tools it may use, whether tool execution is sandboxed, where credentials live and how state is backed up and updated. Sandboxing is off by default; when turned on, tool execution can move into a container or remote host while the Gateway stays on the host.
Security basics
Keep the Gateway on loopback where practical, use SSH or Tailscale for remote access, require Gateway authentication, restrict senders with pairing or allowlists, and treat third-party skills and plugins as code to review before enabling. The OpenClaw docs include a security guide and an exposure runbook.
Where it fits
OpenClaw suits an operator who wants an assistant reachable from chat apps, working on their own machine and accounts. It is not a multi-tenant platform for a whole company's users.
Our service
We deploy OpenClaw on your infrastructure or on an isolated server we manage under a separate agreement, with access, backups and an operating runbook. We are independent and not affiliated with the OpenClaw Foundation.
Keep following the system
Hermes Agent
An open-source, single-tenant personal AI agent from Nous Research that works from a terminal and chat channels.
AI Agent
A system that uses a model and tools to pursue a goal across multiple steps within defined permissions and review boundaries.
Model Context Protocol (MCP)
An open standard for connecting compatible AI applications to tools, resources, and prompts.
Buzz (Block)
An open-source, self-hostable team workspace from Block where people and AI agents share rooms on a signed Nostr event log.
Subagent
A separate AI agent session that a main agent starts for one task, with its own context window, instructions and tools, which returns a short report when done.