Admissions Now Open for our Agentic AI Accelerator Program Register Now
One-Day Course
Master
OpenClaw
in a day.
Go from complete beginner to a fully working personal AI agent — managing your email, calendar, tasks, and automations — in 8 focused, hands-on hours.
Course at a glance
8
Hours of learning
6
Modules
100%
Hands-on
0
Coding required
Beginner friendly
Real workflows
What you'll learn
Not just theory.
Real AI you'll use tomorrow.
Every module is built around activities from your actual day — email, scheduling, research, and automation. By lunch you'll have OpenClaw running on your machine.
Install on your machine
Not a sandboxed demo — your real device, your real data. You leave with a live agent, not a course certificate collecting dust.
Automate your real life
Morning briefings, email triage, weekly reviews, expense capture. Each workflow is chosen because people actually need it.
Build your own skills
Extend your agent with custom skills — no code required. Write a plain Markdown file and your AI gains a new capability.
Security from day one
Privacy and safety aren't an afterthought. You understand what risks exist and how to manage them from the very first session.
Chat via WhatsApp & Telegram
Control your AI from apps already on your phone. No new interfaces to learn — just message it like you'd message a colleague.
No coding background needed
Every technical step is explained in plain English. If you can write an email, you can run this course and build these automations.
Before you arrive
Prerequisites
This course is designed for anyone - no programming experience needed. If you can send a message on WhatsApp, you can use OpenClaw.
What to bring on the day
- A laptop — macOS 12+, Windows 10/11, or Ubuntu 22+
- At least 4 GB of free disk space
- A stable internet connection
- A smartphone with Telegram or WhatsApp installed
- Administrator access to install software
20-minute pre-work
- Install Node.js v24 LTS from nodejs.org
- Create a free account at console.anthropic.com and generate an API key
- Install VS Code (free) or any plain text editor
- Optional but helpful: install Telegram and create an account
No experience needed
You do not need any prior experience with AI tools, APIs, or the command line. The entire course is taught in plain English, and every technical step is guided line by line.
What you'll be able to do
Learning outcomes
Concrete skills you'll walk away with — not vague familiarity.
01
Explain what an AI agent is and how OpenClaw differs from a standard chatbot
02
Install, configure, and validate a working OpenClaw instance on your own machine
03
Connect Telegram, WhatsApp, or Discord to your agent and send real tasks
04
Build a personalised morning briefing that runs automatically at a time they choose
06
Configure a recurring automation using the heartbeat and cron system
07
Identify and apply the top security mitigations to your own setup
08
Monitor API usage, estimate costs, and choose the right model for each task
09
Troubleshoot common issues using built-in tools and community resources
What you build and keep
| What you build | Practical value |
|---|---|
| A working OpenClaw instance paired to Telegram | Your personal AI available 24/7 via the app on your phone |
| Personalised morning briefing | Weather, calendar, emails, and news delivered every morning automatically |
| Email triage command | Inbox summaries with urgency flags, on demand, in seconds |
| Weekly review skill | Guided reflection questions every Friday, answers saved automatically |
| Task nudge automation | A daily reminder if you haven't logged your tasks by 6pm |
| Custom research command | Research reports saved to your Desktop with a single message |
The curriculum
Six modules.
One complete picture.
Every module builds on the last. By the end of Module 02 your agent is live. By Module 05 it's automating your life.
01
What is OpenClaw? The big picture
Concepts, architecture & live demonstration
Learning objectives
- Explain what an AI agent is and how it differs from a chatbot
- Describe the three components of OpenClaw in plain language
- Identify at least three personal use cases before installation begins
- Understand the privacy model: what stays local, what leaves your device
Topics covered
From chatbots to agents
Why OpenClaw is different - it doesn't just respond, it acts. What persistent, proactive AI means in practice.
The OpenClaw architecture
Gateway, messaging platform, and LLM model explained simply. How the three components connect.
Live demonstration
A real-time session: morning briefing, web search, file creation, and a timed reminder - all triggered from Telegram.
Privacy & data ownership
OpenClaw runs locally. What data leaves your device, what doesn't, and how to think about sensitive information.
Choosing your LLM
Claude vs GPT vs local Ollama models. Cost, privacy, and the recommended starting point for beginners.
02
Installation, setup & first message
Hands-on device setup for every participant
Learning objectives
- Install OpenClaw on their own machine (macOS, Windows, or Linux)
- Complete the onboarding wizard and understand each configuration option
- Pair at least one messaging channel and send their first real message
- Read and make simple edits to the openclaw.json config file
- Use the Control UI and /status command to monitor usage and cost
Topics covered
System requirements
Node.js, npm, an API key, and a messaging app - the complete checklist and how to verify each step.
Step-by-step installation
npm install and the onboard wizard, with Windows (WSL2), macOS, and Linux paths covered together.
Pairing your messaging channel
Telegram, WhatsApp, or Discord setup. QR code pairing, verification, and troubleshooting common issues.
Your first conversation
Five tasks to send immediately: date, summary, web search, file creation, and a timed reminder.
openclaw.json config
allowFrom list, model selection, gateway port, and security settings. Plain-English explanations for each.
Control UI & /status
Monitor the agent from the browser. Track API costs, active tasks, and pause or stop the agent safely.
03
Real-world daily workflows
Email, calendar, tasks, research & expense tracking
Learning objectives
- Set up and receive a personalised morning briefing automatically each day
- Use OpenClaw to triage and summarise email from Gmail or Outlook
- Schedule and query calendar events by messaging the agent
- Capture tasks and notes to Notion, Obsidian, or Apple Notes without switching apps
- Run a structured web research task and save the output as a report
Topics covered
Morning briefing
A heartbeat task that sends calendar, top emails, weather, and a headline every morning at your chosen time.
Email triage
'Summarise my inbox from the last 24 hours and flag anything urgent' - live demonstration with a real inbox.
Calendar management
Schedule meetings, find gaps, and set reminders entirely by messaging the agent.
Task & note capture
Capture ideas on the move - they appear instantly in your tool of choice without touching your laptop.
Research on demand tracking
End-to-end: competitors, pricing summaries, and news - formatted report saved to your Desktop automatically.
Expense tracking
Photograph a receipt on your phone, state the category, and have it logged to a spreadsheet automatically.
04
Skills: extending what your agent can do
The skills system, ClawHub & building your own skills
Learning objectives
- Explain what a skill is and how the workspace > global > bundled priority works
- Browse ClawHub and install a community skill via a chat message
- Write a simple custom skill using only a SKILL.md file - no coding required
- Vet a third-party skill for security risks before installing
- Use the meta-skill technique: ask OpenClaw to generate a skill from instructions
Topics covered
What is a skill?
Skill folders, SKILL.md, optional TypeScript, and the priority system explained with examples.
Installing from ClawHub
Browse and install community skills by chatting. Live demo with a popular community-built skill.
Writing your first skill
A mood logging skill from scratch - metadata, trigger phrases, instructions, output format. No code.
Practical skill examples
Bill reminder, grocery list, meeting notes, and weekly review skills. Two built together as a group.
The meta-skill technique
Paste instructions and say 'build a skill from this'. OpenClaw writes its own extension live.
Vetting skills for security
Red flags in SKILL.md: external URL calls, file system access, credential requests. How to sandbox.
05
Automation, proactivity & background tasks
Heartbeats, cron jobs, webhooks & browser automation
Learning objectives
- Explain the difference between a heartbeat, a cron job, and a webhook
- Configure a recurring cron automation for a personal use case
- Wire a webhook to react to an external event without writing code
- Understand the scope and safe limits of browser automation
- Route tasks to different messaging channels by time of day or type
Topics covered
Heartbeats
Configure your agent to reach out proactively - daily digests, bill reminders, meeting alerts.
Cron jobs
Schedule recurring automations. Syntax explained simply. Morning briefing, weekly review, monthly summary.
Webhooks
React to external events: form submissions, GitHub activity, price drops - all without code.
Browser automation
Fill forms, scrape prices, book recurring appointments on your behalf. Scope, limits, and best practices.
Multi-channel routing
WhatsApp for day, Discord for server events, Telegram for personal tasks. Routing in plain English.
06
Safety, privacy & what comes next
Security essentials, cost management & next steps
Learning objectives
- Identify the five most common security mistakes and apply immediate fixes
- Understand prompt injection and how model choice affects risk
- Monitor and control API spending with the built-in tools
- Troubleshoot common problems confidently using logs and Doctor mode
- Leave with a clear set of next actions to keep building momentum
Topics covered
Top 5 security mistakes
Over-privileged skills, exposed gateways, no allowFrom list, unknown skills, and plain-text credentials.
Prompt injection
How a poisoned email can instruct your agent. Why a strong model (Claude) reduces this risk significantly.
Cost management
Light use: $10–30/month. Using /status, switching to local models, and setting hard API usage limits.
Troubleshooting
Reading logs, Doctor mode, community support, and when to restart the agent cleanly.
What comes next
Immediate next actions, recommended community channels, and how to keep building incrementally.
What you'll use
Tools & resources
Every tool used in this course is free. You need no paid subscriptions — just an Anthropic API key (pay-as-you-go, ~$1–2 on the day).
OpenClaw
github.com/openclaw
The agent platform itself. Open-source, runs locally on your device. The core of everything we build.
Free & open-source
Node.js
nodejs.org
The runtime environment for OpenClaw. Install the LTS version before the course.
Free
Anthropic Console
console.anthropic.com
API key generation and usage monitoring. Free account — you only pay for API calls (approx. $1–2 for the full day).
Free account
Telegram
telegram.org
Recommended messaging channel for beginners. Used to send tasks to and receive responses from your agent.
Free
VS Code
code.visualstudio.com
Recommended text editor for editing config files and writing skills. Any plain text editor will work.
Free
ClawHub
Shared during Module 04
The community skills directory. Browse, install, and share skills for OpenClaw — all via a chat message.
Free community
Ready to build your AI?
Enquire about group bookings, corporate delivery, or the next public course date.
Get in touch with Xaltius