Frequently Asked Questions

Common questions about OpenClaw: what it is, how it compares to ChatGPT, installation, cost, local models, security, and where to get help. For more detail, follow the links in each answer or use the OpenClaw Roadmap from the homepage.

Basics

What is OpenClaw?

OpenClaw is a free, open-source autonomous AI agent platform that runs on your own hardware and connects to messaging apps like WhatsApp, Telegram, Discord, Slack, and 15+ others. Unlike cloud chatbots, it doesn't just chat-it acts: it can send emails, run commands, manage files, and automate workflows 24/7. See the full What is OpenClaw? guide.

Who created OpenClaw?

OpenClaw was created by Austrian developer Peter Steinberger. It started as "Clawd" (November 2025), became "Moltbot," then "OpenClaw" (January 2026). In February 2026, Peter announced he was joining OpenAI; the project is transitioning to open-source foundation governance. See History of OpenClaw.

How does OpenClaw work?

A single Gateway process runs on your machine. It receives messages from your connected chat apps (WhatsApp, Telegram, Discord, etc.), sends them to the AI agent, runs skills and tools (email, calendar, shell, etc.), and stores context in a memory system. You control it via the CLI and the Web Control UI in your browser. See Technical architecture for details.

What can OpenClaw do?

OpenClaw can send and read email, manage calendars, run shell commands, control browsers, read and write files, call APIs, and automate workflows 24/7. Use cases include personal productivity (inbox zero, scheduling), business (customer support, sales pipelines), content creation (social media, research), and development (code review, CI/CD alerts). Browse use cases for examples.

OpenClaw vs ChatGPT

Can OpenClaw replace ChatGPT?

It depends on what you need. OpenClaw can replace ChatGPT for many use cases if you want an AI that does tasks (send email, run code, manage calendar) and runs on your own infrastructure. ChatGPT is session-based and doesn't execute actions or connect to your messaging apps. For a detailed comparison, see OpenClaw vs ChatGPT.

Is OpenClaw better than ChatGPT?

OpenClaw is better for privacy (self-hosted), task automation, and using one AI across WhatsApp, Telegram, Discord, etc. ChatGPT can be better for zero-setup chat and polished conversation out of the box. Many users use both: ChatGPT for quick web chat, OpenClaw for automation and messaging. Compare features in our comparison table.

Can I use both OpenClaw and ChatGPT?

Yes. OpenClaw is model-agnostic: you can use Claude, GPT, Gemini, Ollama, or others. You could use OpenAI's API inside OpenClaw and still keep your data on your server and use OpenClaw's automation and channels. There's no requirement to choose one or the other.

Installation & Setup

How do I install OpenClaw?

Quickest: curl -fsSL https://openclaw.ai/install.sh | bash. You need Node.js 22+. Then run openclaw onboard --install-daemon and start the Gateway. See the Quick Start Guide or platform-specific guides: macOS, Windows, Linux, Docker.

What messaging platforms does OpenClaw support?

WhatsApp, Telegram, Discord, Slack, Signal, iMessage, Microsoft Teams, and 15+ others. One Gateway serves all of them. Setup is per channel; see Channel setup (e.g. Telegram, WhatsApp, Discord).

What are the system requirements?

You need Node.js 22+, and a supported OS: macOS, Linux, or Windows (WSL2 recommended on Windows). OpenClaw can run on a local machine, a VPS, or in the cloud (e.g. AWS, DigitalOcean). For resource-heavy local LLMs (Ollama), a capable GPU or sufficient RAM helps. See Installation and Quick start.

Can I run OpenClaw on a VPS or in the cloud?

Yes. Many users run OpenClaw on a VPS (DigitalOcean, Linode, Hetzner, AWS EC2, etc.) or in Docker. See VPS deployment and Docker. Ensure you follow security best practices when exposing or remote-accessing the Gateway.

Cost & Licensing

Is OpenClaw free? What does it cost?

OpenClaw itself is free and open source (MIT). You pay for your own compute (e.g. a VPS or your machine) and, if you use cloud LLMs, for API usage (Anthropic, OpenAI, etc.). If you use local models via Ollama, you can run with no API cost. See our cost calculator for estimates.

Is OpenClaw open source? What license?

Yes. OpenClaw is open source under the MIT License. You can use it commercially, modify it, and redistribute it. There are no usage fees or vendor lock-in from the project itself.

Can I use OpenClaw for business or commercially?

Yes. The MIT license allows commercial use. Many teams use OpenClaw for customer support, sales automation, internal tools, and content workflows. See business use cases and monetization guides if you want to sell services or products around OpenClaw.

Models & Privacy

Can I use OpenClaw with local models (Ollama)?

Yes. OpenClaw supports Ollama and other local providers. You can run fully offline with no data sent to the cloud. Configure in model provider setup (Ollama).

Is OpenClaw safe? What about security?

OpenClaw is powerful-it can run commands and access files-so it must be configured securely. Use our security best practices, network isolation, and sandbox options. Check security overview and advisories before production.

Can OpenClaw access my files?

Yes, by design-so it can read/write documents, run scripts, and automate file-based workflows. You can restrict this with sandbox mode and careful configuration. Never give it more access than you are comfortable with, and follow the security guide.

Why use OpenClaw instead of cloud AI?

Main reasons: data stays on your side (self-hosted), you choose the model (Claude, GPT, Gemini, or local Ollama), real task execution (not just chat), and one agent across many messaging apps. If privacy, control, or automation matters to you, OpenClaw is a strong fit. See OpenClaw vs ChatGPT for a direct comparison.

Skills & Extensions

What are OpenClaw skills? What is ClawHub?

Skills are add-ons that extend what your agent can do (e.g. Gmail, calendar, GitHub, custom APIs). ClawHub is the community marketplace where you discover, install, and publish skills-thousands are available. See the skills marketplace and skills overview.

Can I build my own skills or integrations?

Yes. You can write custom skills (e.g. in TypeScript) or define workflows in YAML/TypeScript. The project supports a plugin/skill architecture; see the official docs and skill development guides. You can keep skills private or publish them to ClawHub.

Alternatives & Comparisons

What are OpenClaw alternatives?

Alternatives depend on what you need: cloud chat AI (ChatGPT, Claude web), workflow automation (n8n, Make, Zapier), or multi-agent frameworks (AutoGen, CrewAI). OpenClaw is unique in combining self-hosted AI + messaging channels + task execution. See comparisons: vs ChatGPT, vs n8n, best AI agents.

How does OpenClaw compare to n8n or other automation tools?

n8n and Make are great for visual, trigger-based workflows. OpenClaw is conversational and agent-driven: you talk to it in natural language and it decides and executes steps. It also connects directly to WhatsApp, Telegram, etc. You can use both-e.g. n8n for deterministic pipelines, OpenClaw for chat-driven automation. Details: OpenClaw vs n8n.

Troubleshooting

OpenClaw command not found / Gateway won't start-what do I do?

Ensure Node.js 22+ is installed and the openclaw CLI is on your PATH (the installer usually does this). If the Gateway won't start, check ports, config file (~/.openclaw/openclaw.json), and logs. Use openclaw doctor if available. Full steps: troubleshooting hub, installation errors, gateway issues.

Where do I report bugs or get technical support?

Report bugs and request features on GitHub Issues. For real-time help, ask in the Discord community. OpenClaw Roadmap’s troubleshooting and error code reference can also help.

Getting Help

Where can I get help?

Join the Discord community, search or open an issue on GitHub, or use our troubleshooting hub. For guides and the full path from install to production, start at the OpenClaw Roadmap homepage.