Frequently Asked Questions

ClawdiusAI is a hosted multi-agent AI platform built on OpenClaw. You define a mission — like "grow my SaaS to $10K MRR" or "manage my content pipeline" — and ClawdiusAI breaks it into tasks, delegates to specialist AI agents, and drives everything to completion. Each user gets a dedicated server, visual dashboard with kanban board, and an AI Orchestrator that coordinates the entire team.

OpenClaw is an open-source AI agent framework. It lets you run autonomous agents — each with their own personality, skills, LLM model, and communication channels. Each agent runs as a persistent "gateway" process. ClawdiusAI is a fully managed hosting platform for OpenClaw. We handle server provisioning, configuration, SSL, DNS, updates, and monitoring. You get all the power of OpenClaw without any infrastructure work.

ChatGPT, Claude, and Gemini are single AI assistants — you chat with one model, one conversation at a time. They don't have persistent identity, they don't run tasks in the background, and they can't coordinate with other agents. ClawdiusAI gives you a team of specialized agents that work together autonomously. A content writer, SEO analyst, developer, and research assistant can all work in parallel on your mission, coordinated by an AI Orchestrator. The agents are always running — they don't wait for your next message to act.

CrewAI, AutoGen, and LangGraph are multi-agent frameworks — they help developers build agent systems in code. They're libraries, not products. You still need to write the orchestration logic, set up infrastructure, and build a UI. ClawdiusAI is a ready-to-use platform. You don't write code — you define a mission in plain English, create agents through a wizard, and manage everything through a visual dashboard. The orchestration, task routing, and agent communication are all built in.

No. When you sign up and pay, ClawdiusAI automatically provisions a dedicated cloud server for you. It comes pre-configured with OpenClaw, the dashboard, Caddy (for SSL), and your Orchestrator agent. Everything is ready in under 60 seconds. You access it through a browser at yourname.cloud.clawdiusai.com.

BYOK means you can use your own LLM API keys — from OpenRouter (recommended, gives you access to all providers), or direct from Anthropic, OpenAI, etc. This gives you full control over costs and model selection. Alternatively, you can use our managed keys for zero-friction onboarding: you don't create any external accounts, we handle everything, and token usage is tracked in your dashboard. You can switch between BYOK and managed keys anytime.

Yes. ClawdiusAI supports webhook-based channel integrations. The most common setup: connect your Orchestrator to Telegram so it can send you daily summaries, alert you about blockers, and receive instructions from you on the go. WhatsApp and other messaging platforms work similarly. The dashboard's built-in activity feed is always available as the primary communication channel.

Completely. Every ClawdiusAI account runs on its own dedicated virtual machine — not a shared container, not a multi-tenant database. Your data, your agents, your API keys, and your conversations never touch anyone else's infrastructure. If you delete your account, the entire server is destroyed.

ClawdiusAI itself — the control plane, provisioning, billing, and orchestration layer — is not open source. However, the AI agent framework underneath is OpenClaw, which is open source. If you want to self-host and build everything yourself, you can use OpenClaw directly. ClawdiusAI is for people who want the managed, batteries-included experience.

ClawdiusAI supports all major LLM providers through OpenRouter: Claude (Anthropic), GPT-4 and GPT-4o (OpenAI), Gemini (Google), Llama (Meta), Mistral, and many others. Each agent can use a different model and even a different provider. If you're on BYOK, you can use any provider that OpenRouter supports, or connect directly to Anthropic/OpenAI.

The base plan gives you a cloud VM with 2 vCPUs and 2GB RAM — enough for an Orchestrator plus 3-4 specialist agents. If you need more capacity, we automatically migrate you to a larger server (4GB, 8GB) with zero downtime. Server costs are included in your plan — no surprise infrastructure bills.

During beta, server access is dashboard-only. We're evaluating SSH access for power users who want to customize their OpenClaw configuration beyond what the wizard offers. If this is important to you, let us know.