Skip to Content
View all events

AI Superstream: OpenClaw and Friends

Published by O'Reilly Media, Inc.

Beginner to advanced content levelBeginner to advanced

Creating the future of agentic assistants and AI workflows with "Claws"

In the last three months, AI assistance has moved beyond simple tasks like editing and coding into real assistance. Reasoning and agents—connected to new tools like OpenClaw, NanoClaw, and OpenFang—turn chatbots into persistent, determined extensions of ourselves. This frontier is evolving so rapidly it’s hard to keep track. Every week, there’s a new skill, a new agent, and a new scandal. But it’s vital to separate the signal from the noise, because agentic AI threatens to upend not only business and technology, but the economy and society as a whole.

What better way to understand this new frontier than to watch these agents, and the people behind them, in real time? Join us for a high-energy showcase of next-generation digital AI agents designed to think, act, and execute like a true extension of you. Experience live demonstrations of AI assistants that manage workflows, automate complex tasks, integrate with your favorite tools, and unlock new levels of productivity. Engage with industry leaders to evaluate the impact of AI on professional workflows and gain a technical understanding of how autonomous assistants are transitioning from simple tools to transformative agents. Stay ahead of the curve by understanding the applications, potentials, and intricacies of AI personal assistants, and seeing how they're transforming the way we work.

What you’ll learn and how you can apply it:

  • Build reliable AI-driven projects by translating intent into clean, scalable code using agentic assistants
  • Scale your content engine into a cross-platform powerhouse, fueling social growth through automated, intelligent workflows
  • Secure your agentic workflows by transforming risky AI actions into auditable, human-in-the-loop successes
  • Deliver secure, client-facing agentic solutions by leveraging flexible AI assistants

Recommended Resources

Schedule

The time frames are only estimates and may vary according to how the class is progressing.

Introduction (5 minutes)

Alistair welcomes you to the AI Superstream.

Securing Claw Agents Where It Counts: The Execution Layer – Eran Sandler (25 minutes)

OpenClaw, NanoClaw, and other claw instances are built to do real work—but that also means they can cause real damage. Eran Sandler, founder of Canyon Road and creator of AgentSH, demonstrates how to protect claw-based agents at the execution layer, the critical boundary where model intent becomes file writes, process launches, and network access. You’ll see a live demo of claw instances under both normal and adversarial conditions, including prompt injection and unsafe tool behavior. You’ll also watch AgentSH intercept runtime activity to enforce deterministic policy: allow, deny, or require human approval. By proving that prompt guardrails alone are insufficient, this talk illustrates how to build a last line of defense that keeps autonomous systems safer and more accountable, even when models make mistakes.

Trust, but Verify: Building and Auditing OpenClaw Skills – Kesha Williams (25 minutes)

Skills are playbooks for your OpenClaw agent. They tell it how to handle a specific job, which tools to use, and what steps to take. But most people install them from a marketplace without ever reading the playbook first. Kesha Williams, founder and principal AI architect at Keysoft and an AWS AI Hero, takes you through the full skill lifecycle: building one cleanly, testing that it behaves as expected, and deploying it without giving your agent more access than it needs. She also audits a third-party skill live, showing what to look for, including hidden instructions, file system reach, network calls, and tool permissions you didn’t know you granted. You’ll leave with a checklist you can run against any skill before it touches your environment.

5 Things to Do After Installing OpenClaw (Before You Break It) – Erik Hanchett (25 minutes)

Most people install OpenClaw, get excited and skip the boring setup, then quit before the second week. They don't realize their gateway is exposed to the public internet or that their default model is costing them five times what it should or that their agent has no boundaries on what it can do without asking. Dev advocate and Program With Erik host Erik Hanchett takes you through the five things that actually matter in your first week, from cost control to security to getting your agent to stop doing things you didn't ask for, all in a live Lightsail instance so you can see for yourself.

Break (5 minutes)

Engineering with AI Assistants: From Prompt Chaos to Reliable Systems – Ari Joury (25 minutes)

Ari Joury, the CEO of Wangari Global, demonstrates how to evolve from simple code-completion to a “collaborative partner” model, where AI assistants function as integrated engineering peers. You’ll learn practical patterns for transforming vague prompts into rigorous specifications and how lightweight evaluation loops can prevent architectural decay in machine-generated projects. Using NanoClaw and others in the agentic ecosystem, Ari also shows you how to scaffold high-integrity workflows that remain readable and testable, ensuring that even as machine-generated output scales, the resulting systems stay strictly aligned with human engineering intent.

One Hour, 20 Pieces: How AI Agents Turned My Live Stream Into a Content Machine – Kyle Balmer (25 minutes)

Kyle Balmer, creator of AI with Kyle, demonstrates how to produce over 20 pieces of daily content using a custom “useful layer” agentic swarm. Leveraging the same systems he uses to reach a global audience of over 250,000 followers, Kyle illustrates how Claude Cowork and other agentic assistants can handle everything from overnight research briefings to the automated transformation of raw transcripts into polished newsletters and SEO-optimized web stories. Through this high-speed “human-to-agent” handoff, you’ll understand that AI doesn’t replace the creator; rather it builds an automated pipeline that allows leaders to scale their message and authority across multiple simultaneous platforms at the speed of use.

Closing Remarks (5 minutes)

Alistair closes out today’s event.

Your Hosts and Selected Speakers

  • Alistair Croll

    Alistair Croll is an entrepreneur, author, and conference organizer. He's written four books on technology and society, including the best-selling Lean Analytics, which has been translated into eight languages. He's the cofounder of web performance startup Coradiant (acquired by BMC), the Year One Labs startup accelerator, and a number of other early-stage companies.

    A prolific speaker, Alistair was a visiting executive at Harvard Business School, where he helped create a course on data science and critical thinking. He's founded and chaired a number of the world's leading technology events including Cloud Connect, Strata, Startupfest, Scaletech, and the FWD50 Digital Government conference. He's currently working on Just Evil Enough, the subversive marketing playbook. Alistair lives in Montreal, Canada, and writes at acroll.substack.com.

  • Eran Sandler

    Eran Sandler is co-founder and CEO of Canyon Road, the company behind AgentSH, an execution-layer security platform for AI agents. He is a longtime entrepreneur, engineer, and security leader who has built and led multiple startups across security, identity, and fintech. His work today focuses on making agentic systems safer in the real world.

  • Ari Joury

    Ari Joury is a machine learning engineer and CEO of Wangari Global, working at the intersection of AI, decision-making systems, and real-world production environments. With a background spanning theoretical physics, applied ML, and enterprise analytics, Ari’s focus is on building AI systems that move from prediction to action—and helping teams integrate AI assistants into serious engineering workflows.

  • Kyle Balmer

    Kyle Balmer is the founder of AI with Kyle, a daily AI education brand that has reached over 300,000 entrepreneurs and business owners with practical AI education. He’s spent 20 years building businesses, including Vietnam’s first private TV station, and publishing eight best-selling Amazon books. Through his daily livestream, newsletter, and social content, he’s become one of the most recognized independent voices in AI education, translating AI from hype into plain-English practical skills for nontechnical business owners. His mission: getting one million people AI-ready. Kyle has a BA in history from Oxford and an MBA from NYU Stern.

  • Erik Hanchett

    Erik Hanchett is an author, developer, and teacher who's spent the last 10+ years helping people learn to code on his YouTube channel, Program With Erik (100K+ subscribers). He’s also a senior developer advocate at AWS. Erik loves speaking at conferences around the world about frontend development.

  • Kesha Williams

    Kesha Williams is an enterprise architect and AI consultant with over three decades of experience designing, building, and operating production software systems. She’s the founder and managing partner of Keysoft, where she helps organizations safely adopt and operationalize AI, including autonomous agent systems. An AWS AI Hero, Kesha focuses on helping engineering and security teams deploy AI safely by applying practical architectural and operational patterns. She brings a practitioner’s perspective shaped by experience working directly with teams deploying agent-based systems in production. Kesha is also an AI educator and advisor, as well as a frequent speaker at global technology conferences.

    linkedinXlinksearch

Skill covered

Generative AI