OpenClaw: How to Actually Use it, with Practical Instructions

[1-hour] The first wave of AI agents gave us chatbots that could answer questions. Now we're entering the era of autonomous AI agents that actually do things — managing your email, triaging GitHub issues, monitoring your infrastructure, and running workflows 24/7 while you sleep. But most people deploying these agents are stuck fighting Docker configs, managing infrastructure, and debugging integrations instead of focusing on what the agent should actually be doing. KiloClaw eliminates that entire layer. This workshop teaches you how to deploy, configure, and get real value from your own fully-managed OpenClaw agent on Kilo's infrastructure.

Brian Turcotte

Developer Relations - Kilo Code

Developer Relations Engineer @ Kilo Code, where we're building the most complete coding agent that eliminates friction and lets engineers focus on what matters. On the content side, I facilitate product marketing launches, educational videos, technical walkthroughs, and written/visual content across our blog, social media, and website. I also collaborate with AI model partners on promotional events and work to improve our documentation and technical resources. On the community side, I facilitate live webinars, discussions, and meetups that bring together Kilo users and the broader open-source community. It's one of the most rewarding parts of the job - connecting with developers who are pushing the boundaries, and learning from how they're using our tools in the real world.

Workshop Overview

Part 1 — What KiloClaw Actually Is (and Why It Matters)

OpenClaw in 60 seconds: the open-source autonomous agent with 350k+ GitHub stars that connects to your real tools and takes real actions
The self-hosting trap: why running your own OpenClaw instance is a time sink most people don't need
KiloClaw's value prop: one-click provisioning, managed infrastructure, automatic API key management, no Docker configuration required
What's in the box: headless Chromium, GitHub CLI, Google Workspace integration, 500+ model access via Kilo Gateway
Live demo: Spinning up a KiloClaw instance from scratch and having it respond on Telegram in under 2 minutes

Part 2 — Practical Use Cases You Can Implement Today

Developer workflow automation: Connect GitHub, set up a bot account with scoped permissions, and let KiloClaw triage issues, open PRs, and leave code reviews autonomously
Email and calendar management: Connect a dedicated Google account and let KiloClaw draft replies, schedule meetings, and surface what matters from your inbox
Business operations: Lead research, CRM updates, prospect outreach, and web monitoring with the browser tool and OpenClaw skills system
Personal productivity: RSS monitoring with Blogwatcher, automated summaries of long content, cross-platform messaging coordination across Telegram, Discord, and Slack
Live demo: Setting up GitHub integration with a fine-grained PAT, Google Workspace via the Docker auth flow, and a Telegram channel — then watching KiloClaw handle a real multi-step workflow across all three

Part 3 — Configuring, Securing, and Scaling Your Agent

The skills system: how SKILL.md files work, installing from ClawHub, and writing your own custom skills
Model selection strategy: when to use Claude vs. GPT vs. DeepSeek vs. free models depending on task complexity and cost
Security best practices: dedicated accounts, scoped tokens, understanding the "full" tool profile and what your agent can access
Prompt patterns that actually work: giving your agent clear objectives vs. vague instructions, and how to structure multi-step workflows
Hands-on exercise: Participants configure their own KiloClaw instance with at least one integration (GitHub, Google, or chat platform) and deploy a custom skill

Key Takeaways

From Chatbot to Agent: Understand the difference between an AI that answers questions and an AI that autonomously executes workflows across your real tools and services.
Practical Playbook: Leave with a working KiloClaw instance connected to your actual tools, with at least one automated workflow running.
Build Once, Run 24/7: Learn how to set up an agent that works for you around the clock — handling tasks across GitHub, Google Workspace, and messaging platforms without ongoing babysitting.

Requirements:
Laptop with a web browser; a Kilo account (free to create at kilo.ai); optionally, a dedicated GitHub account and/or Google account for integrations. No local development environment needed — KiloClaw runs entirely in the cloud.

OpenClaw: How to Actually Use it, with Practical Instructions

Brian Turcotte

Developer Relations - Kilo Code

Workshop Overview

Part 1 — What KiloClaw Actually Is (and Why It Matters)

OpenClaw in 60 seconds: the open-source autonomous agent with 350k+ GitHub stars that connects to your real tools and takes real actions

The self-hosting trap: why running your own OpenClaw instance is a time sink most people don't need

KiloClaw's value prop: one-click provisioning, managed infrastructure, automatic API key management, no Docker configuration required

What's in the box: headless Chromium, GitHub CLI, Google Workspace integration, 500+ model access via Kilo Gateway

Live demo: Spinning up a KiloClaw instance from scratch and having it respond on Telegram in under 2 minutes

Part 2 — Practical Use Cases You Can Implement Today

Developer workflow automation: Connect GitHub, set up a bot account with scoped permissions, and let KiloClaw triage issues, open PRs, and leave code reviews autonomously

Email and calendar management: Connect a dedicated Google account and let KiloClaw draft replies, schedule meetings, and surface what matters from your inbox

Business operations: Lead research, CRM updates, prospect outreach, and web monitoring with the browser tool and OpenClaw skills system

Personal productivity: RSS monitoring with Blogwatcher, automated summaries of long content, cross-platform messaging coordination across Telegram, Discord, and Slack

Live demo: Setting up GitHub integration with a fine-grained PAT, Google Workspace via the Docker auth flow, and a Telegram channel — then watching KiloClaw handle a real multi-step workflow across all three

Part 3 — Configuring, Securing, and Scaling Your Agent

The skills system: how SKILL.md files work, installing from ClawHub, and writing your own custom skills

Model selection strategy: when to use Claude vs. GPT vs. DeepSeek vs. free models depending on task complexity and cost

Security best practices: dedicated accounts, scoped tokens, understanding the "full" tool profile and what your agent can access

Prompt patterns that actually work: giving your agent clear objectives vs. vague instructions, and how to structure multi-step workflows

Hands-on exercise: Participants configure their own KiloClaw instance with at least one integration (GitHub, Google, or chat platform) and deploy a custom skill

Key Takeaways

From Chatbot to Agent: Understand the difference between an AI that answers questions and an AI that autonomously executes workflows across your real tools and services.

Practical Playbook: Leave with a working KiloClaw instance connected to your actual tools, with at least one automated workflow running.

Build Once, Run 24/7: Learn how to set up an agent that works for you around the clock — handling tasks across GitHub, Google Workspace, and messaging platforms without ongoing babysitting.

Requirements:
Laptop with a web browser; a Kilo account (free to create at kilo.ai); optionally, a dedicated GitHub account and/or Google account for integrations. No local development environment needed — KiloClaw runs entirely in the cloud.

Time and Location

March 31, 2026

11:30pm - 12:30pm

Cobb Galleria

Workshop Requirements

OpenClaw: How to Actually Use it, with Practical Instructions

Brian Turcotte

Developer Relations - Kilo Code

Workshop Overview

Part 1 — What KiloClaw Actually Is (and Why It Matters)

OpenClaw in 60 seconds: the open-source autonomous agent with 350k+ GitHub stars that connects to your real tools and takes real actions

The self-hosting trap: why running your own OpenClaw instance is a time sink most people don't need

KiloClaw's value prop: one-click provisioning, managed infrastructure, automatic API key management, no Docker configuration required

What's in the box: headless Chromium, GitHub CLI, Google Workspace integration, 500+ model access via Kilo Gateway

Live demo: Spinning up a KiloClaw instance from scratch and having it respond on Telegram in under 2 minutes

Part 2 — Practical Use Cases You Can Implement Today

Developer workflow automation: Connect GitHub, set up a bot account with scoped permissions, and let KiloClaw triage issues, open PRs, and leave code reviews autonomously

Email and calendar management: Connect a dedicated Google account and let KiloClaw draft replies, schedule meetings, and surface what matters from your inbox

Business operations: Lead research, CRM updates, prospect outreach, and web monitoring with the browser tool and OpenClaw skills system

Personal productivity: RSS monitoring with Blogwatcher, automated summaries of long content, cross-platform messaging coordination across Telegram, Discord, and Slack

Live demo: Setting up GitHub integration with a fine-grained PAT, Google Workspace via the Docker auth flow, and a Telegram channel — then watching KiloClaw handle a real multi-step workflow across all three

Part 3 — Configuring, Securing, and Scaling Your Agent

The skills system: how SKILL.md files work, installing from ClawHub, and writing your own custom skills

Model selection strategy: when to use Claude vs. GPT vs. DeepSeek vs. free models depending on task complexity and cost

Security best practices: dedicated accounts, scoped tokens, understanding the "full" tool profile and what your agent can access

Prompt patterns that actually work: giving your agent clear objectives vs. vague instructions, and how to structure multi-step workflows

Hands-on exercise: Participants configure their own KiloClaw instance with at least one integration (GitHub, Google, or chat platform) and deploy a custom skill

Key Takeaways

From Chatbot to Agent: Understand the difference between an AI that answers questions and an AI that autonomously executes workflows across your real tools and services.

Practical Playbook: Leave with a working KiloClaw instance connected to your actual tools, with at least one automated workflow running.

​Build Once, Run 24/7: Learn how to set up an agent that works for you around the clock — handling tasks across GitHub, Google Workspace, and messaging platforms without ongoing babysitting.

Requirements: Laptop with a web browser; a Kilo account (free to create at kilo.ai); optionally, a dedicated GitHub account and/or Google account for integrations. No local development environment needed — KiloClaw runs entirely in the cloud.

Time and Location

March 31, 2026

11:30pm - 12:30pm

Cobb Galleria

Workshop Requirements

Build Once, Run 24/7: Learn how to set up an agent that works for you around the clock — handling tasks across GitHub, Google Workspace, and messaging platforms without ongoing babysitting.

Requirements:
Laptop with a web browser; a Kilo account (free to create at kilo.ai); optionally, a dedicated GitHub account and/or Google account for integrations. No local development environment needed — KiloClaw runs entirely in the cloud.