Frequently asked questions
Real answers to common questions about EasyClaw and OpenClaw.
No. With EasyClaw you do not install OpenClaw. EasyClaw runs OpenClaw for you. You sign up, connect Telegram, choose your model (Claude, ChatGPT, or Gemini), and start chatting. There is no Docker, no servers, and no command-line setup.
EasyClaw is a managed platform that runs OpenClaw. OpenClaw is the open-source agent runtime. EasyClaw hosts it, connects it to Telegram, and handles billing and configuration so you can use OpenClaw without self-hosting.
You can stop using EasyClaw and self-host OpenClaw at any time. OpenClaw is open source; your agent logic and use cases are not locked in. EasyClaw does not prevent you from moving to your own deployment later.
Yes. We use Row Level Security (RLS) for your data, isolated execution per agent, and encrypted storage. Your agent’s memory and context are not shared with other users. We do not log message content for marketing. Security is managed by us so you don’t have to configure servers or firewalls.
No. When you use EasyClaw with our default models, we run the model for you. You choose Claude, ChatGPT, or Gemini in the dashboard; no API key is required from you. Usage is included per plan.
Telegram only for the MVP. It’s the fastest path to a working agent with zero setup. Other channels may be added later.
We run a shared OpenClaw runtime with strict agent-level isolation. Your agent’s memory and execution are fully separated from other users. You do not get a dedicated VM per account; you get a dedicated logical agent and session.
No. Each agent runs in an isolated namespace. Browser sessions used for tasks are ephemeral and unique to each run. No data or cookies are shared between users.