How to Set Up OpenClaw with VidClaw: Complete Guide

In this guide, you'll set up an OpenClaw AI agent on your own VPS and connect it to VidClaw — a self-hosted dashboard for managing tasks, tracking usage, and shaping your agent's behavior. By the end, you'll have a fully working AI agent with a visual command center you can access from anywhere.

What is OpenClaw?

OpenClaw is an open-source framework for running autonomous AI agents. It gives your agent a workspace on your server where it can execute tasks, write code, manage files, and interact with APIs. Think of it as the engine that powers your AI agent — it handles scheduling, memory, tool use, and model communication.

OpenClaw runs headlessly on your server. You configure it, point it at a model provider (like Anthropic), and it works autonomously on tasks you assign.

What is VidClaw?

VidClaw is a self-hosted dashboard that sits on top of OpenClaw. It gives you a visual interface to manage everything your agent does:

  • Kanban board — queue and prioritize tasks with drag-and-drop
  • Usage tracking — real-time token counts and cost estimates
  • Skills manager — enable, disable, or create agent skills
  • Soul editor — shape your agent's persona and instructions
  • Activity calendar — see what your agent worked on each day
  • File browser — browse your agent's workspace files

VidClaw binds to localhost only — you access it securely through an SSH tunnel. No accounts, no cloud, no tracking.

Prerequisites

Before you start, make sure you have:

  • A VPS — any Linux server (Ubuntu 22.04+ recommended) with at least 1 GB RAM
  • SSH access — you should be able to SSH into your server as root or a sudo user
  • An Anthropic API key — sign up at console.anthropic.com and create an API key
  • Basic terminal skills — you'll be running commands on your server

Setup Steps

Prepare your VPS

SSH into your server and make sure everything is up to date:

ssh root@your-server-ip

# Update packages
apt update && apt upgrade -y

# Install required dependencies
apt install -y git curl

You'll also need Node.js 18+. If it's not installed:

curl -fsSL https://deb.nodesource.com/setup_20.x | bash -
apt install -y nodejs

Install OpenClaw

Clone the OpenClaw repository and run the installer:

git clone https://github.com/openclaw/openclaw.git ~/.openclaw
cd ~/.openclaw
./install.sh

During installation, you'll be prompted for your Anthropic API key. Paste it when asked.

Verify the installation:

openclaw --version

Tip: OpenClaw stores its configuration at ~/.openclaw/openclaw.json. You can edit this file directly, or use VidClaw's Settings panel later.

Configure your agent

Set up your agent's workspace and persona. OpenClaw uses a workspace directory where your agent stores its files, memory, and task data:

mkdir -p ~/.openclaw/workspace

Edit the OpenClaw config to set your preferred model and workspace path:

# ~/.openclaw/openclaw.json
{
  "model": "claude-sonnet-4-20250514",
  "workspace": "~/.openclaw/workspace",
  "api_key": "sk-ant..."
}

Note: You can switch models at any time from VidClaw's navbar — no need to edit this file manually once VidClaw is running.

Install VidClaw

The fastest way to install VidClaw is the one-liner:

curl -fsSL vidclaw.com/install.sh | bash

This installs dependencies, builds the frontend, and sets up a systemd service automatically.

Alternatively, install manually:

cd ~/.openclaw/workspace
git clone https://github.com/madrzak/vidclaw.git dashboard
cd dashboard
./setup.sh

The setup script handles everything: npm dependencies, Vite build, and systemd service registration.

Access the dashboard

VidClaw binds to localhost:3333 for security. To access it from your local machine, create an SSH tunnel:

ssh -L 3333:localhost:3333 root@your-server-ip

Now open http://localhost:3333 in your browser. You should see the VidClaw dashboard with the Kanban board ready to go.

Tip: Add an alias to your shell config for quick access:

alias vc="ssh -L 3333:localhost:3333 root@your-server-ip"

Create your first task

With the dashboard open, it's time to give your agent something to do:

  1. Click the + Add button at the top of the Todo column
  2. Enter a task title — for example, "Write a README for the project"
  3. Set a priority (P1 is highest) and optionally attach a skill
  4. Click Create — your task appears on the board

Your OpenClaw agent picks up tasks from the Todo column automatically, ordered by priority. As it works, the card moves to In Progress, and when it's done, it lands in Done with a timestamp.

You can also use the quick-add bar at the top of the Kanban board. Type your task, use @skill to attach a skill, and hit Enter.

Next Steps

You now have a fully working OpenClaw agent with VidClaw as your command center. Here are some things to explore next:

  • Explore all features — usage tracking, skills manager, soul editor, and more
  • Read the install guide — configuration options, updating, and uninstalling
  • Customize your agent's soul — open the Soul Editor panel and pick a starter template
  • Set up recurring tasks — use the schedule feature to automate repetitive work
  • Star the repo — if VidClaw is useful, a star helps others find it

Ready to take control of your AI agent?

VidClaw is free, open-source, and takes 60 seconds to install.