OpenClaw is an autonomous open-source AI agent (MIT License) that runs locally as a Node.js service on your own infrastructure. Unlike traditional chatbots, OpenClaw executes real-world tasks autonomously: managing emails, organizing calendars, controlling browsers, writing code, editing files and much more. Its unique heartbeat system checks for pending tasks at configurable intervals and acts proactively – without requiring a prompt every time.
Trusted by leading companies
OpenClaw is an autonomous open-source AI agent (MIT License) that runs locally as a Node.js service on your own infrastructure. Unlike traditional chatbots, OpenClaw executes real-world tasks autonomously: managing emails, organizing calendars, controlling browsers, writing code, editing files and much more. Its unique heartbeat system checks for pending tasks at configurable intervals and acts proactively – without requiring a prompt every time.
With persistent SQLite-based memory, OpenClaw remembers context, preferences and work habits across sessions. Over 100 preconfigured integrations (Gmail, Calendar, GitHub, Notion, Spotify and more) and multi-channel communication via WhatsApp, Telegram, Slack, Discord, Signal, iMessage, Microsoft Teams, Matrix and WebChat make OpenClaw the most versatile self-hosted AI agent on the market. The TypeScript-based project was created by Peter Steinberger (PSPDFKit founder) and has over 170,000 GitHub stars.
We install, host and operate OpenClaw for your company – either on our secure, GDPR-compliant infrastructure in Germany or on-premise in your own environment. Leverage the benefits of an autonomous AI agent without vendor lock-in.
With 24/7 monitoring, enterprise support, automated backups and professional maintenance, we ensure maximum availability and reliable operation of your OpenClaw instance. Including assistance with integrating your preferred LLM providers and messaging channels.
The unique heartbeat system checks every X minutes (configurable, default: 30) for pending tasks. OpenClaw reads a HEARTBEAT.md file, detects action items and executes tasks autonomously – checking flight status, sending reminders, summarizing conversations – without an explicit prompt.
SQLite-based memory preserves context, preferences and history across sessions. OpenClaw remembers your work habits, project details and personal settings – becoming more useful over time. Semantic search enables fast retrieval of previous interactions.
Communicate with OpenClaw via your preferred channel: WhatsApp, Telegram, Slack, Discord, Signal, iMessage, Microsoft Teams, Matrix, Google Chat or WebChat. A central WebSocket gateway coordinates all clients seamlessly.
Gmail, Calendar, GitHub, Notion, Spotify, browser automation, file system, terminal and over 100 more preconfigured AgentSkills. OpenClaw executes real tasks in your favorite tools – and can teach itself new skills when needed.
OpenClaw runs entirely on your own infrastructure as a Node.js service. Your data, conversations and automations never leave your environment. Perfect for GDPR-compliant companies, government agencies and regulated industries. You choose the LLM provider.
Fully open source under MIT license with over 170,000 GitHub stars. Complete transparency, independent security audits and unlimited customization possibilities. No vendor lock-in – the agent belongs to you.
Open source enterprise-ready for productive workloads - we run your applications with highest security standards and enterprise support
Open source software for business-critical processes requires professional maintenance, continuous updates, and enterprise-grade support. With our OpenClaw Enterprise Managed Hosting, you get the necessary infrastructure and support to reliably operate open source in production environments. Backups, SLAs, telephone support, and personal contact - so you can focus on your core business.
We also offer customized OpenClaw Enterprise solutions for your specific requirements. Contact us for an individual quote.
From fully managed GPU servers to compact AI Cubes – we provide the ideal infrastructure for your local LLM applications.
Powerful GPU servers with dedicated hardware for compute-intensive LLM workloads. Fully managed, scalable, and optimized for maximum performance.
Compact AI workstation for local LLM inference. Perfect for office environments, with top-tier performance and absolute data sovereignty.
Good choice – we'll help you get started or with operations.
As a Managed Service customer at WZ-IT, you have access to our exclusive portal: Monitor your infrastructure in real-time, schedule maintenance, request quotes, and get direct support – all in one central location.
Answers to the most important questions
Topics
OpenClaw is an autonomous open-source AI agent that runs locally as a Node.js service and executes real-world tasks independently. Unlike chatbots such as ChatGPT, OpenClaw can manage emails, organize calendars, control browsers, write code, edit files and operate over 100 other tools. The project is under MIT license and was created by Peter Steinberger.
OpenClaw was created by Peter Steinberger, an Austrian software engineer and founder of PSPDFKit (PDF developer tools). The project started as a weekend experiment and went viral, growing to over 170,000 GitHub stars. OpenClaw is fully open source under MIT license and is actively developed by a vibrant community.
ChatGPT and other chatbots only respond to prompts and cannot execute real actions. OpenClaw, on the other hand, is an autonomous agent with a heartbeat system that proactively handles tasks, has persistent memory and can actually interact with your tools through 100+ integrations – sending emails, creating appointments, running code, editing files, controlling browsers and more. OpenClaw uses LLMs (Claude, GPT, DeepSeek etc.) as its 'brain' but is itself the 'hands'.
OpenClaw requires Node.js >= 22 and runs on macOS, Linux and Windows (via WSL2). Installation is done via npm as a global package using the 'openclaw onboard' command. A WebSocket gateway runs on port 18789 and coordinates all clients. Conversation data is stored as JSONL files under ~/.openclaw/, memory uses SQLite.
OpenClaw is installed via npm and provides a guided onboarding wizard ('openclaw onboard'). Alternatively, it can be deployed via Docker. For professional installations with SSO, VPN access and security hardening, we offer a complete installation service – on-premise or in the cloud.
Yes, OpenClaw is fully self-hostable. You can run it on your own server, in your cloud infrastructure (Hetzner, AWS, Azure, GCP) or on-premise. We support you with installation, configuration and ongoing operations on your preferred infrastructure.
OpenClaw supports WhatsApp (via Baileys), Telegram (via grammY), Slack, Discord (via discord.js), Signal, iMessage, Microsoft Teams, Matrix, Google Chat and WebChat. Through the WebSocket gateway, all channels can be used simultaneously – a message via WhatsApp triggers the same agent as a Slack command.
OpenClaw offers over 100 preconfigured AgentSkills: Gmail, Google Calendar, GitHub, Notion, Spotify, browser automation (Semantic Snapshots), file system, terminal/SSH, code execution and more. New integrations can be developed as TypeScript modules – or the agent can teach itself new skills when needed.
Yes, OpenClaw is written in TypeScript (~40,000 lines of code) and is fully extensible. You can develop custom AgentSkills as TypeScript modules, customize existing skills or ask the agent to build the necessary integration itself when accessing new data sources.
OpenClaw itself does not send data to third parties. Conversation data is stored locally in JSONL files, memory uses a local SQLite database. However, prompts are sent to the configured LLM provider (Claude, OpenAI, DeepSeek etc.) – when running locally with Ollama, even this data stays within your infrastructure.
When OpenClaw runs on infrastructure in a German data center and uses a local LLM provider (e.g., Ollama), no personal data leaves the system – full GDPR compliance. When using cloud LLMs, the data protection impact assessment should consider token transmission. Our managed hosting is fully GDPR compliant.
The heartbeat system is OpenClaw's unique feature for proactive autonomy. At configurable intervals (default: 30 minutes), the gateway sends a heartbeat prompt to the agent. The agent reads a HEARTBEAT.md file in the workspace, checks for pending tasks and either acts (sends messages, executes tasks) or stays silent (HEARTBEAT_OK). This way, OpenClaw independently monitors situations and acts without manual input.
OpenClaw uses a local SQLite database for semantic search across all previous interactions. Conversations are stored as JSONL files under ~/.openclaw/. The agent remembers preferences, work habits, project context and previous instructions – becoming more useful and personalized over time.
Yes, we offer fully managed OpenClaw hosting on GDPR-compliant infrastructure in Germany. Including 24/7 monitoring, automated backups, security updates, SLA-based support and assistance with integrating your LLM providers and messaging channels. Optionally with a dedicated GPU server for local LLM inference.
OpenClaw uses a lane queue system for serial task execution that prevents race conditions. For teams, multiple OpenClaw instances can be run in parallel – each with their own workspaces, memory and channel connections. We assist with the architecture and operation of multi-instance setups for enterprise environments.
More questions? We are happy to help!
These solutions are often used together with OpenClaw
These solutions offer similar functionalities and can be evaluated together
Whether a specific IT challenge or just an idea – we look forward to the exchange. In a brief conversation, we'll evaluate together if and how your project fits with WZ-IT.
Trusted by leading companies
Timo Wevelsiep & Robin Zins
CEOs of WZ-IT
