One section in Chapter 2 really resonated with me. Stulberg calls it selecting worthwhile pursuits — and it gave me a diagnostic I am starting to use on every new project, every inbound opportunity, every commitment I’m weighing.

TL;DR:

• Decades of research on motivation point to three core needs: autonomy, competence, and belonging

• Before saying yes to a project, job, or commitment, ask: will this increase or decrease autonomy, competence, or belonging in my life?

• In an AI-disrupted career, this filter matters more — because the wrong yes compounds faster

• For existing commitments, ask what conversations or moves would restore any of the three

• Bonus: turn the filter into a prompt and riff with Claude or ChatGPT on the decisions you’re stuck on

The Three Needs

Stulberg pulls from decades of self-determination research. We thrive over the long haul when three needs are met:

Autonomy — some control over how we spend our time and energy. Not total freedom. Just a real say in the how.

Competence — a path toward concrete improvement in our chosen pursuits. The thing we’re doing needs to grow our craft, not just consume our hours.

Belonging — connection to something beyond ourselves. A person, a community, a lineage, a tradition. An emotional stake in why this work exists.

His line that stopped me: “The more time and energy we spend on pursuits that afford us autonomy, competence, and belonging, the better.”

Simple. But almost nobody applies it when evaluating the next thing on their plate.

Why This Filter Matters Right Now

A year ago, a lot of work was good enough. The pay was fine, the scope was clear, the path was predictable. You could stay on autopilot and be okay.

That’s over. AI is redrawing the edges of what humans should do. Tasks that felt worth doing in 2024 are now commodity output. Roles that felt stable are getting reassembled. The leaders I coach are making harder choices — about what to say yes to, what to drop, and what to rebuild.

In that environment, the wrong yes doesn’t just waste time. It compounds. Every month spent on a pursuit that drains autonomy, stalls competence, or isolates you is a month your peers are getting sharper on the things that will actually matter in 2027.

Which is why Stulberg’s question has become my default diagnostic:

Will this increase or decrease autonomy, competence, or belonging in my life?

How I’m Using It

Three ways I’ve been running this filter in the last few weeks:

On new opportunities. When something lands in my inbox — a speaking gig, a consulting ask, a collaboration — I don’t jump to the calendar. I ask the three. Does it give me more control over my time, or chain me to someone else’s schedule? Does it push my craft forward, or just repeat what I already know? Does it connect me to people and ideas I care about, or is it a transaction?

If two of the three are a no, I pass. No agonizing.

On existing commitments. Stulberg adds a second move most people skip. For work you’re already in, ask what actions would enrich the three. How can I protect more autonomy here? What conversation with my team would rebuild competence momentum? What would reconnect me to why this work matters? A good commitment doesn’t need to be abandoned — it often just needs a small intervention.

As a prompt, when I’m stuck. This is where the AI angle gets practical. When I’m genuinely torn, I paste the situation into Claude and run it through the filter:

I'm deciding whether to [take on / continue / walk away from] this project:

[brief description — scope, time commitment, stakeholders, outcome]

Use Brad Stulberg's framework from The Way of Excellence.
Evaluate this on three dimensions:
1. Autonomy — will it increase or decrease my control over my time and energy?
2. Competence — will it grow my craft, or just consume hours?
3. Belonging — does it connect me to something beyond myself, or is it purely transactional?

Give me a direct read on each, then the trade-offs I should weigh before saying yes or no.

Claude doesn’t decide for me. But it surfaces the angles I’d missed — especially the ones I was quietly avoiding.

The Harder Version of the Question

The sharpest line in the chapter is this: “What would it look like to shape our lives for more mastery and mattering?”

That’s the harder question. Not should I say yes to this one thing, but:

Am I designing my life around pursuits that grow me and connect me — or am I letting it happen to me?

In an AI era where almost everything about work is in motion, the leaders I watch thriving are the ones asking this out loud. They’re not waiting for the dust to settle. They’re picking the pursuits that compound autonomy, competence, and belonging — and pruning the ones that don’t.

That’s the real work. AI is the accelerant. But what you’re accelerating — that’s still the decision only you can make.

Your Turn

Pick one commitment on your plate right now. Run it through the three.

→ Is it giving you more or less autonomy than it did six months ago?
→ Is it building real competence, or just repeating what you already do well?
→ Do you feel connected to the people and the purpose behind it?

If the answer is honest and uncomfortable, that’s the signal. The next move isn’t always quit — sometimes it’s a conversation, a scope change, a recommitment. But the filter cuts through the noise.

Reply and tell me what came up. I read every response.

Stay curious. Stay hands-on.

James

The book is The Way of Excellence by Brad Stulberg. Worth reading the whole chapter on the psychology of excellence — chapter two is where this framework lives.

If this was useful, forward it to one person weighing a hard yes right now. They can subscribe at graymatter.jamesgray.ai.

We’d spent weeks learning tools, building workflows, writing prompts. And then, one by one, people started sharing — not what they’d built, but who they were becoming. Someone said AI had made them more creative than they’d felt in years. Another said they could see a broader version of who they could be. A third said: for the first time in a long time, I know what I want to do next.

That wasn’t a conversation about AI. That was a conversation about the Life’s Task.

Robert Greene writes about this in Mastery — the idea that each of us has a primal inclination, a thread that runs through everything, and that the work of a lifetime is to find it and follow it fully. I’ve been reading Greene for years. But I didn’t expect AI to be the thing that brought his theory to life in real time — accelerating the moment when people finally see what’s possible for them.

AI isn’t just changing how we work. It’s forcing a deeper question: what are you actually pointing this at?

Subscribe now

Five Strategies for Finding Your Life’s Task

Greene offers five strategies. I walked through all of them in this episode — here’s the map.

1. Return to Your Origins — The Primal Inclination Strategy
Go back before the job titles, before the expectations, before the world told you what to be good at. What drew you in as a child? What did you lose yourself in? I ask this in every cohort, and what strikes me is how quickly people remember — and how long they’ve been ignoring it. That pull was never random. It was always pointing somewhere.

2. Occupy the Perfect Niche — The Darwinian Strategy
Don’t compete in a crowded lane. Find the intersection that is uniquely yours — where your combination of skills, experience, and inclination has no direct competition. That’s where you thrive.

3. Avoid the False Path — The Rebellion Strategy
Some paths are chosen for the wrong reasons — money, approval, inertia. Greene calls this the false path. Recognizing it takes courage. Leaving it takes more.

4. Let Go of the Past — The Adaptation Strategy
What got you here won’t necessarily get you there. If AI has reshaped your industry or your role, the task isn’t to hold on — it’s to find what carries forward and build from there.

5. Find Your Way Back — The Life-or-Death Strategy
Some people only find their Life’s Task after a crisis forces the question. A job loss. A health scare. An industry upended. The disruption isn’t the end — it’s the redirection.

The Inner Quest

The Inner Quest is a series within the Graymatter podcast — dedicated to one of its three pillars: mastering yourself. Alongside mastering AI and building what matters, this is the thread I believe matters most.

Not tools. Not frameworks. The deeper journey — to find ourselves, evolve ourselves, and adapt ourselves. The quest that runs beneath everything else. The one that doesn’t end.

Every episode, one idea worth sitting with. This is that series.

Your Reflection Prompt

Somewhere inside you, you already know.

There is something — in your heart, in your bones — that is your Life’s Task. Something that would bring out your uniqueness in a way nothing else can. You’ve caught glimpses of it. You may have pushed it away. It’s likely that you haven’t had the courage to fully touch it, to say it out loud, to pursue it.

That’s not weakness. That’s human. The Life’s Task asks everything of you, and that’s terrifying.

But here’s what I want you to sit with: we don’t know how many days we have. None of us do. And when you hold that truth — really hold it — the question changes.

Not “what should I do with my career?”

But: what would I honor?

What is the one thing you can see, right here, right now — your Life’s Task, your opportunity, the thing that is uniquely yours to bring into the world?

Write it down. Even one sentence. That’s where it begins.

I’d love to know what comes up for you. Drop it in the comments — even one sentence. You might be surprised what you write.

-James

Most AI education gives you pieces. A prompting trick here. A tool demo there. Disconnected fragments that leave you with the same question you started with:

How do I actually go from my business process to a working AI system?

Over the next three weeks, I’m running three free Lightning Lessons on Maven that answer that question — start to finish.

Each session is 30 minutes. Each one builds on the last. Together, they walk you through the Business-First AI Framework — the same 7-step methodology I teach in two cohort courses — from deconstructing a messy process to testing and iterating a production-ready AI workflow.

Here’s what’s coming.

Subscribe now

Week 1: Deconstruct (Framework Step 2)

March 27 | 30 min | Free on Maven

Your 5-step process is actually 15 steps. You just can’t see them yet.

That’s the core insight of the Deconstruct phase. Most AI projects fail not because the technology breaks — but because the process wasn’t broken down enough for AI to execute reliably. Hidden decisions, implicit data handoffs, undocumented context needs, and failure modes only surface after you’ve built the wrong thing.

In this session, I’ll deconstruct a real business process live using a 5-question framework:

• Discrete Steps — What are the actual sub-steps hiding inside each “step”?

• Decision Points — Where does someone make a judgment call?

• Data Flows — What goes in, what comes out, where does it come from?

• Context Needs — What templates, docs, or references does each step require?

• Failure Modes — What does bad output look like, and how would you catch it?

By the end, a simple-looking process becomes a fully mapped workflow definition — the kind of clarity AI actually needs to do useful work.

You’ll walk away with: A repeatable methodology you can apply to any process in your organization. No AI tools required — just sharper thinking about how work actually works.

🎟️ Register for free on Maven

Week 2: Design + Build (Framework Steps 3-4)

March 31 | 30 min | Free on Maven

You’ve deconstructed the process. Now what?

This session picks up where Week 1 leaves off. Starting from a workflow definition, I’ll walk through the Design and Build phases — live on Claude.

First, Design: you’ll see how each step in a deconstructed process maps to AI building blocks. Which steps become prompts, which become skills, which need agent orchestration, which require integrations. The design matrix classifies your workflow — deterministic, guided, or autonomous — and that classification constrains what you can build.

Then, Build: you’ll watch the actual construction. A workflow definition becoming executable artifacts in Claude Code, in real time. The definition drives every decision. You’re not guessing what to create — the Deconstruct output tells you exactly what artifacts you need.

The key insight: Build is step 4 of 7. Shipping the first version isn’t the finish line. Test and Improve are what turn a build into a production system — and that’s where we’re headed in Week 3.

You’ll walk away with: A clear mental model for turning any deconstructed process into a working AI system — and an understanding of what comes next to make it production-ready.

🎟️ Register for free on Maven

Week 3: Test + Iterate (Framework Steps 5 & 7)

April 3 | 30 min | Free on Maven

You’ve built the workflow. Now prove it works — and make it better.

This is where most people stop. They ship v1 and hope for the best. But the Business-First AI Framework has two more phases for a reason: Test verifies your build produces consistent, quality output across real scenarios. Improve optimizes for speed, cost, and reliability so the workflow actually gets used in production.

In this session, I’ll show both phases in action — using Claude skill configuration as the lens:

• Structured evals (Test) — Run evaluations that verify your skills produce consistent output before you rely on them. Evals catch the inconsistencies that spot-checking misses.

• Effort levels (Improve) — Control how deeply Claude thinks before responding. High for nuanced analysis, low for straightforward transformations. Match the compute to the job.

• Model selection (Improve) — Match Opus, Sonnet, or Haiku to each skill based on the reasoning it requires. Not every skill needs the most powerful model.

The punchline: a well-configured skill on Haiku can outperform a poorly configured one on Opus. Testing tells you which configuration actually works. Iteration makes it cheaper and faster without sacrificing quality.

You’ll walk away with: An eval methodology to verify your builds work — and a configuration matrix for optimizing quality, speed, and cost across any Claude skill.

🎟️ Register for free on Maven

Why These Three Sessions Fit Together

This isn’t three random topics. It’s one framework:

Analyze → Deconstruct → Design → Build → Test → Run → Improve

That’s the Business-First AI Framework — the 7-step methodology I teach in my Maven courses and at Berkeley. These three Lightning Lessons cover the critical phases:

• Week 1 (Deconstruct) exposes the real complexity in your process

• Week 2 (Design + Build) turns that clarity into a working AI system

• Week 3 (Test + Improve) proves the system works and makes it faster, cheaper, and more reliable

Most people skip straight to building. Then they wonder why the results are inconsistent. The reason is almost always the same: the process wasn’t broken down enough, or the system was never tested and tuned for the job.

These sessions give you the key phases in 30 minutes each. My Maven courses — Hands-on Agentic AI for Leaders and Claude for Builders — give you the full lifecycle with hands-on practice on your own workflows.

All Three Are Free. All Three Are Live.

Live demos, real processes, real builds. And if you can’t make it live, register anyway. You’ll get the recording and the slide deck after each session.

Register for any or all:

• March 27: Deconstruct Any Business Process into an AI Workflow

• March 31: Build an AI Workflow from a Business Process

• April 3: Configure Claude Skills for Quality, Speed, and Cost

If you know someone who’s been trying to figure out how to go from business process to working AI, forward this to them. These three sessions cover the complete path.

Stay curious. Stay hands-on. 🙌

James

You set up an AI workflow to handle your weekly pipeline review. It’s supposed to pull deals from your CRM, flag anything stalled, draft a summary, and post it to Slack. You kick it off and step away.

Three minutes later: “Do you want to allow this Notion query?” You approve it. Two minutes later: “Do you want to allow this Slack message?” You approve it. Then: “Do you want to allow this CRM update?” Three more approvals follow in the next five minutes.

By the time the workflow finishes, you’ve approved eleven actions. A task that was supposed to run while you did something else pulled you out of focus seven times. You might as well have just done it yourself.

This is the approval-fatigue problem with agentic AI that nobody mentions in the demos. And until last week, Claude Code gave you exactly two ways to handle it — neither of them designed for real business work.

TL;DR

• Claude Code shipped auto-mode on March 24 — a new permissions setting between “approve everything” and “skip everything”

• Before each tool call (Notion, Slack, Gmail, CRM, Sheets — any MCP connection), a classifier decides if it’s safe to proceed automatically or risky enough to block

• Low-risk actions execute without interrupting you; potentially destructive actions get blocked before they run

• This is the thing that makes business workflows actually autonomous — not just automated with a human stuck in the approval loop

• It’s a research preview for Team plan users now; Enterprise and API users get it in the coming days

• Honest limitation: the classifier isn’t perfect, and Anthropic still recommends isolated environments for now

The All-or-Nothing Problem (Before Auto-Mode)

If you’re running Claude Code with MCP connections to your business tools — Notion, Slack, Gmail, HubSpot, Google Sheets — you’ve been living with a frustrating binary.

Option A: Default mode. Every MCP tool call asks for approval. Claude wants to read a Notion database? Approve. Write a new record? Approve. Send a Slack message? Approve. Update a deal stage in your CRM? Approve. The workflow is safe. It’s also not autonomous in any meaningful sense. You’re the approval layer, which means you’re still doing the work — just with extra steps.

Option B: --dangerously-skip-permissions. No approvals, ever. Claude runs from start to finish without checking in. This sounds like the autonomous workflow you wanted until you think about what you’ve traded: any ability to catch a mistake before it executes. Mass-delete the wrong Notion records? Send a draft Slack message to the wrong channel? Write junk data to a production CRM? You find out after the fact.

The flag is named --dangerously-skip-permissions for a reason. Anthropic isn’t being dramatic. For workflows touching real business data and real communication channels, skipping permissions entirely is a bad bet.

Neither option is actually right for how you want to use Claude Code as a business operations tool. Default mode is too slow. Skip-permissions is too reckless. The gap between them is where the real frustration lives.

What Auto-Mode Actually Does

Anthropic shipped auto-mode on March 24 as a research preview for Team plan users. The mechanism is straightforward: before each tool call runs, a classifier reviews it.

The classifier categorizes the action — safe or risky — and routes accordingly:

• Safe actions proceed automatically. Reading records, writing standard updates, and posting messages within normal parameters. Claude keeps moving without interrupting you.

• Risky actions get blocked. The classifier flags potential problems — bulk deletes, data exports to unexpected destinations, actions that appear destructive — and redirects Claude to find an alternative approach.

• Persistent blocks surface a prompt. If Claude keeps attempting actions that keep getting blocked, it eventually surfaces a permission request to you. The system doesn’t silently fail — it escalates when it actually needs to.

That last point matters. Auto-mode isn’t binary. It’s a dial. The classifier handles the routine. You get pulled in when there’s genuinely something to decide.

For a business operations workflow, this changes the picture entirely. A pipeline review that used to ping you 11 times now runs to completion with 1 or 2 meaningful check-ins. A daily brief that pulls from Notion, summarizes it, and posts to Slack actually executes while you’re in your first meeting of the day. The approvals that do come through are the ones that actually warrant your attention.

What This Changes for Business Operations

The operations workflows I see leaders building — and wanting to build — are exactly the category auto-mode unlocks.

Think about the kinds of tasks that make sense to delegate to Claude Code with MCP connections:

• Pull data from a CRM or spreadsheet, identify exceptions, draft a report, and post it to a Slack channel

• Monitor an inbox, triage inbound requests, update a Notion tracker, and draft responses for review

• Scan a project management board, flag overdue items, update statuses, and send a summary to stakeholders

• Reconcile meeting notes against a Notion database, extract action items, and create task records

Every one of these workflows involves multiple sequential tool calls. In default mode, each call is an interruption. That makes the workflow feel more like a demo than a deployment — something you’d show to impress someone, not something you’d actually rely on to do work while you’re busy doing other work.

Auto-mode is what makes these workflows real. Not impressive. Real.

The permission-fatigue problem isn’t just about convenience — it’s about what happens to your attention when approvals are required for everything. When you have to approve every low-stakes action, you stop carefully reviewing the approvals. Everything becomes a quick click. And that’s the moment you’re most likely to miss the one approval that actually matters.

Matching the approval friction to the actual risk of the action is how you build a system where human judgment shows up where it’s worth something.

The Junior Operator Analogy

Here’s the mental model I keep coming back to.

Think about how a strong operations coordinator handles delegated work. They handle the standard decisions — updating the tracker, sending the routine check-in, filing the document where it belongs — without asking you every time. That’s not because they’re unsupervised. It’s because the trust level for those decisions is already established.

But they bring you in on things that are outside their lane: a message that doesn’t have an obvious right answer, an action that could be hard to undo, a decision where they genuinely aren’t sure what you’d want. The judgment about what to escalate is itself a skill — and a good coordinator gets it right most of the time.

Auto-mode is Claude Code developing that judgment for MCP tool calls. It’s not unlimited autonomy. It’s calibrated autonomy, matched to the actual risk of each individual action. The routine proceeds. The genuinely uncertain gets flagged.

That’s the working relationship you want with an AI system doing operations work — not “check with me constantly” and not “never check with me.” Something in between, calibrated to what actually warrants a decision.

What Auto-Mode Can’t Do (Being Honest)

Auto-mode is a research preview. That’s not a disclaimer to skip — it matters for how you use it.

The classifier isn’t perfect. Anthropic says this directly: some risky actions will slip through when intent is ambiguous, or when Claude doesn’t have enough context about your environment to recognize additional risk. Some benign actions will also get blocked. The system is good, but calibrate your expectations accordingly.

It reduces risk compared to --dangerously-skip-permissions — it doesn’t eliminate it. Anthropic explicitly recommends using auto-mode in isolated environments. If you’re running it against production Notion workspaces, live CRM data, or active communication channels — the calculus hasn’t shifted as much as you might think. Start with test environments, read-only connections, or low-stakes workflows while you build a feel for how the classifier behaves.

There’s a small cost. The classifier runs before each tool call, resulting in additional token consumption and marginal latency. For most business workflows, this is negligible. Worth knowing it exists.

Admins can disable it. For team deployments, this is important: administrators can turn off auto-mode entirely with "disableAutoMode": "disable" in managed settings. If you’re planning to roll this into team workflows, check with whoever manages your Claude Code deployment first.

The Bigger Pattern: Autonomy Is a Dial, Not a Switch

Here’s the practitioner take that applies well beyond auto-mode.

Most business leaders think about AI autonomy as a binary: either you’re in control, or the AI is. Either you approve everything, or you trust it blindly. That framing produces exactly the two bad options we started with.

The more useful mental model: autonomy is a dial, calibrated to the actual risk of individual decisions. Some decisions are low-stakes and should proceed automatically. Others warrant human judgment. A well-designed system knows the difference.

Auto-mode is the first serious implementation of this in Claude Code. And it points to a principle that should shape every AI workflow you build or oversee.

When you’re connecting Claude Code to your business tools via MCP — Notion, Slack, Gmail, HubSpot, Google Sheets, whatever your stack looks like — the autonomy question is always present. How much should the agent decide on its own? When should it check in? What constitutes an action that actually requires a human?

The wrong answer is “always check in” (you’re the bottleneck, there’s no leverage) or “never check in” (the risk is unmanaged, the accountability is gone). The right answer is: match autonomy to actual decision risk. High-frequency, low-stakes tool calls proceed automatically. Novel situations, irreversible actions, and genuinely ambiguous contexts surface human judgment.

Auto-mode operationalizes this for MCP tool calls. Your job, as someone building or managing AI-connected business workflows, is to operationalize it across your entire workflow design. That’s not a technical question — it’s a judgment call about which decisions actually require you.

How to Get Started

Auto-mode is available now as a research preview for Claude Team plan users. Enterprise and API users will get access in the coming days.

To enable it from the CLI:

claude --enable-auto-mode

Then cycle to it using Shift+Tab within a session.

On Desktop or in the VS Code extension: Auto-mode is disabled by default on the desktop app, so you’ll need to enable it first. Go to Settings → Claude Code and toggle auto-mode on. Once enabled, you can select it from the permission mode dropdown within any session.

Works with: Claude Sonnet 4.6 and Opus 4.6.

My recommendation: start with a workflow with low stakes and a clear scope. A Notion query that reads but doesn’t write. A summary workflow that drafts a Slack message for your review instead of sending it directly. Get a feel for what the classifier flags and what it passes. Build your own calibration for when to trust it before you put it in front of anything consequential.

That’s the same principle I teach in my agentic AI work: don’t assume the system’s judgment is right for your context. Observe it, then extend trust incrementally.

If you’re building business operations workflows with Claude — and you want a structured way to set up skills, agents, and MCP integrations that actually run your business — that’s exactly what we build in Claude for Builders. The next cohort starts on April 13th.

Stay curious. Stay hands-on.

James

Hannah Pritchett, Chief People Officer at Anthropic, spent six months trying to explain to her team what she wanted them to build. She described it in meetings. She wrote it down. They came back with what they thought they heard. None of it was right.

So over a weekend, she built a rough prototype herself using Claude Code. She showed it to her team. Their response: “Now I understand what you want.”

Six months of misalignment. Resolved the moment she stopped describing and started building.

TL;DR:

• AI executes your ambiguity perfectly — if you can’t articulate what you want, AI won’t figure it out for you

• The two leadership skills that matter most in the AI era: clarity (knowing what to ask for) and discernment (knowing whether you got it)

• You can only develop these skills by building — not by delegating, reviewing decks, or attending demos

• 86% of C-suite leaders are increasing AI investment, but only 20% of employees feel like co-creators in how it changes their work

• The leaders who close that gap are the ones who build alongside their teams — not above them

The Clarity Gap Is a Leadership Gap

Melissa Daimler wrote a piece in Forbes this week that landed for me because she put a name on the exact leader I’ve been teaching for the last two years — at UC Berkeley Executive Education, at Maven, and across every cohort I’ve run. The builder-leader. The leader who gets hands-on before asking others to build.

Her core argument: the biggest problem with AI in organizations isn’t the technology. It’s that leaders haven’t done the hard work of getting clear about what they actually want. And her sharpest line: “AI will execute your ambiguity perfectly.”

Over the last year, I’ve watched this identity emerge in real time across my Maven cohorts. CEOs, entrepreneurs, founders, SVPs, marketing leaders — people who didn’t come from technical backgrounds but who chose to get hands-on anyway. They embody this builder-leader persona: they learn by doing, they build before they delegate, and they develop sharper instincts because of it.

I’ve also met the other kind. Leaders who view hands-on work as beneath them. Who believe their role is to set direction and let technical people figure out the details. A year ago, that was a defensible position. In 2026, those leaders are becoming irrelevant — and fast.

The pattern is always the same. A leader assigns a deliverable. An employee uses AI to produce it. The leader asks one clarifying question. The employee can’t answer it. The leader blames the tools.

But the tools aren’t the problem. The prompt was vague going in. The output was accepted without scrutiny on the way out.

Here’s the harder question Daimler asks — one most leaders don’t want to hear: Could you have answered that clarifying question yourself? Had you ever done the work you just assigned?

If not, you don’t just have a clarity problem with your employee. You have one with yourself.

What Builder-Leaders Do Differently

The builder-leader isn’t a leader who codes for a living. It’s a leader who has done the work at least once, so they know what good looks like. That distinction changes everything about how you lead an AI-augmented team.

A leader who hasn’t built asks: “Why doesn’t this look right?”

A builder-leader asks: “What context did you give it? How specific were your prompts? What did you do when the first output wasn’t right?”

That second conversation is a coaching moment. The first is a dead end.

This is what I’ve been building my entire teaching practice around. When executives in my courses sit down with Claude Code and build something — a workflow, a prototype, a decision tool — the shift is immediate. They stop asking “what can AI do?” and start asking “what do I actually need?” Those are completely different questions, and only the second one leads anywhere useful.

Daimler’s Paris travel app story makes this tangible. She built an app to organize hundreds of restaurant and café recommendations. The first version was functional but flat. Then she got clearer: organize by arrondissement so her family could use it contextually. Then clearer still: design it so her family would actually want to open it. Each round of building forced a new round of clarity. The prototype wasn’t the product. The prototype was the clarity.

The Two Skills AI Demands But Can’t Give You

Daimler identifies two skills that separate builder-leaders from everyone else: clarity and discernment.

Clarity is the front end. What you ask for, how you scope it, what quality looks like before AI generates anything. This is harder than it sounds. So much of what makes a great leader lives in their head — instincts, pattern recognition, the “I just know.” AI forces you to make that implicit knowledge explicit. You cannot prompt what you cannot articulate.

Discernment is the back end. The judgment to evaluate whether AI’s output is actually the outcome you wanted. AI produces output confidently. It will write you a board memo that sounds perfect — clean structure, polished prose — while the market data is from 2023, the competitor analysis includes a company that was acquired last year, and the recommendation contradicts your actual strategy.

Discernment is catching that before it goes out. And you can only catch it if you’ve done the work yourself at least once.

This maps directly to what I see in my courses. I developed a framework called Business-First AI that walks leaders through three steps: Analyze the workflows that matter, Deconstruct them into the detail required to build, and Build the actual solution.

Step two — Deconstruct — is where the a-ha happens. Every time.

Leaders walk in thinking they understand their workflows. Then I give them 30 minutes to break one down: every decision point, every input, every handoff, every edge case.

When we reconvene, the reactions are always the same. They’re amazed by the edge cases they’d never considered, the details they’d forgotten, the steps they’d been carrying in their heads without ever making explicit. They leave with a new appreciation for the level of clarity truly required for a machine to carry out the workflow that lives in their head.

That’s the clarity gap Daimler is writing about, experienced in real time.

By the time they reach step three — Build — the hard work is already done. The prototype comes together faster because the thinking is sharper. The clarity came from the deconstruction, not from the tool.

You can’t develop taste by reading a menu. You develop it by cooking.

What’s at Stake

Here’s what leaders are risking by staying hands-off.

Your credibility is at stake. BCG found that when leaders demonstrate strong support for AI, the share of employees who feel positive about it rises from 15% to 55%. But “support” isn’t sending a company-wide email or mandating adoption. It’s leaders who are experimenting with the tools, working through the same challenges, and building alongside their teams. Your employees know the difference between a leader who has built something and one who’s read the executive summary.

Your strategic vision is at stake. If you haven’t used AI to build, you’re forming strategy based on vendor demos and consultant decks. You’re making the biggest workforce decisions in a generation — investing and cutting — without the firsthand understanding those decisions require. Accenture’s 2026 data is stark: 86% of C-suite leaders are increasing AI investment, but only 20% of employees feel like active co-creators. That gap doesn’t close with better tools. It closes when leaders develop the clarity to know what they want and the discernment to know when they’ve got it.

Your ability to lead is at stake. The leader who hasn’t built can’t coach their team through AI challenges. They can’t ask the right follow-up questions. They can’t distinguish between AI slop and genuine insight. And here’s the uncomfortable truth Daimler surfaces: “AI slop” isn’t the AI’s fault. Human slop is accepting it. If you can’t tell the difference, that’s not a technology problem. That’s a leadership problem.

Your competitive position is at stake. The organizations moving fastest with AI aren’t the ones with the biggest budgets. They’re the ones where leaders have built enough to articulate a clear vision, set specific quality standards, and recognize when output meets them. Every week you wait to get hands-on is a week your competitors are developing the clarity and discernment you’re still outsourcing.

If you think getting hands-on is beneath you, you’re wrong. It’s beneath you to lead a team through a transformation you haven’t experienced yourself.

Start Here

Pick one deliverable you’ve been about to assign. A report, a strategy doc, an analysis. Do it yourself first. Take 30 minutes. Open Claude, ChatGPT, or whatever tool your company uses, and start with a real problem you’ve been putting off.

Be specific going in. Don’t say “give me an analysis of our metrics.” Say “I want a metrics report that tells me the story behind the numbers, identifies the top two or three gaps we need to address, and ends with agreed-upon next actions we can bring to the team.” Upload relevant documents. Give examples of what good looks like. Set constraints.

Evaluate what comes back with real critical thinking. AI doesn’t signal uncertainty. It sounds confident whether it’s right or wrong. That’s your job to catch.

Then bring what you learned back to your team. Show them what you built. That conversation — where you share your process, your prompts, your mistakes — is the work. That’s how you become a builder-leader.

Want a structured way to start? I built a Business-First AI plugin for Claude Code that guides you through all three steps of the framework — Analyze, Deconstruct, and Build. Install the plugin and it walks you through identifying the right workflow, breaking it down to the level of clarity AI requires, and constructing the solution. It’s the same process my cohort students use, available as a set of skills you can run on your own.

You won’t get it right the first time. As Daimler puts it: “The question is whether you’ll get the Eiffel Tower or the Egyptian pyramids. And whether you’ll know the difference.”

Master AI. Master Yourself. Build What Matters.

This article reinforced something I believe deeply: the AI era doesn’t reward the leaders who adopt the fastest. It rewards the leaders who get clear the fastest. And clarity only comes from building.

The great leaders I work with aren’t just learning AI. They’re fusing AI with their own leadership identity — reimagining what it means to set vision, to coach teams, to make decisions. They see the opportunity and they’re embracing it, reshaping who they are as leaders at the intersection of technology and judgment.

That’s the real work. Not mastering AI as a tool. Mastering yourself as a leader — and building what matters from that foundation. The two are inseparable now.

Stay curious. Stay hands-on.

-James

Going deeper. If this resonates and you want to go from reading about builder-leadership to practicing it, my Maven courses are built for exactly that. Claude for Builders teaches you to build real workflows with Claude Code and Cowork. Hands-on Agentic AI for Leaders takes you from building blocks to full AI-powered business workflows. Both have April cohorts open for enrollment. No slides. No theory. You build.

TL;DR:

• There are 10 building blocks organized into 3 layers: Intelligence, Orchestration, and Integration

• Most workflows only need 2-4 blocks — you don’t need all ten

• The starter combo is Prompt + Context. Add blocks as complexity demands

• Skills teach HOW. Agents EXECUTE. Together, they’re the orchestration duo

• This framework gives you a shared vocabulary to design, discuss, and improve every AI workflow you touch

• Deep-dive reference for each block: Agentic AI Building Blocks on the Hands-on AI Cookbook

Why a Framework Matters

Here’s the problem I see across executive education cohorts: teams jump straight into tools without a shared vocabulary. One person says “agent” and means a chatbot. Another means an autonomous system. A third means a Project workspace with uploaded files.

The result? Misaligned expectations, over-engineered solutions, and workflows that break when the person who built them goes on vacation.

Across 6,000+ professionals I’ve taught, the organizations that move fastest with AI share one trait: they have a common language for what they’re building. These 10 building blocks are that language.

One more thing before we dive in. The blocks are tools — but the person wielding them determines the outcome. A framework without judgment produces noise. Judgment without a framework wastes time. The goal is both.

The Three Layers

The building blocks stack into three layers. Each layer builds on the one below it.

Intelligence (foundation) — the persistent engine, knowledge, and workspace powering every interaction.

Orchestration (middle) — the instructions, routines, and autonomous agents that direct and do the work.

Integration (top) — the protocols, interfaces, and frameworks that bridge AI to external systems and code.

Think of it like a building. Intelligence is the foundation and utilities. Orchestration is the people and processes inside. Integration is the doors, windows, and connections to the outside world.

Let’s walk through each block with real examples you can apply this week.

Layer 1: Intelligence (The Foundation)

These four blocks are the persistent infrastructure beneath every AI interaction.

1. Model

The AI engine — a system trained on data that takes input and produces output through learned patterns.

Choosing the wrong model is like hiring a senior architect to answer the front desk phone. You waste money on overkill or get poor results from underkill. The key decision is always speed vs. depth vs. cost.

Use a fast model (Claude Haiku, Gemini Flash) to auto-tag 500 support tickets overnight — cheap and fast. Use a reasoning model (ChatGPT, Claude Opus, Gemini Pro) to analyze patterns across those tickets and recommend process changes — slower and more expensive, but worth it for strategic insight.

Try this: Pick one recurring task and ask — am I using the right model tier? Most people default to the most powerful model for everything. That’s like taking a helicopter to the grocery store.

2. Context

This is the one that separates useful AI from impressive-but-useless AI.

Context is the unique knowledge your workflow requires that isn’t baked into the model — your files, docs, databases, proprietary data. It comes in three forms: uploaded files (PDFs, spreadsheets), referenced documents (Google Docs, Notion pages), and connected data (databases, live systems).

Here’s the test: “Analyze recurring themes in customer feedback and suggest which product feature to prioritize next quarter.” Generic prompt, generic answer. Now add your last three quarterly reports as uploaded PDFs. Completely different output — grounded in your reality, not the model’s training data.

I see this gap constantly in executive education. People complain that AI gives them “generic” answers. Nine times out of ten, they haven’t given it anything specific to work with. Context is the fix.

Try this: If you’re trying to find customers for your business, write up your buyer persona in a markdown file — who they are, what they care about, what problems they’re solving. Now attach that file any time you ask AI to draft outreach, write copy, or brainstorm campaigns. Same file, every time, dramatically better output.

3. Memory

Memory is what turns AI from a stranger into a colleague who knows how you work.

After several sessions, your AI assistant remembers that you structure reports as executive summaries. It knows you prefer bullet points over paragraphs. It recalls that your team uses OKRs, not KPIs. You didn’t configure any of this. It learned from working with you. Claude, ChatGPT, and Gemini all support memory — the implementations differ, but the concept is the same.

People confuse Memory and Context constantly. Here’s the distinction: you curate context (you choose which files to upload). The system curates memory (it observes your patterns and stores them automatically). One is your briefcase. The other is institutional knowledge.

Try this: Use the same AI tool for 3-4 sessions this week. Then ask: “What have you learned about my preferences?” The answer reveals how memory compounds with use. And here’s a trick — you don’t have to wait for the AI to pick up on patterns organically. You can tell it directly: “Remember that my team reports on a quarterly cadence” or “Remember that I prefer concise executive summaries over detailed narratives.” That fact gets stored in memory and applied in future sessions.

4. Project

If you’ve ever started a new AI conversation and spent the first five minutes re-explaining who you are and how you work — you need a Project.

Projects are self-contained workspaces with their own chat histories, knowledge bases, and custom instructions. Think of them as folders that actively shape the AI’s behavior. Create a “Q4 Investor Updates” project, upload your financial data and competitive landscape docs, set the tone and format rules once — and every conversation in that workspace inherits it all.

The killer feature is isolation. Your investor update project doesn’t leak into your marketing project. Different rules, different context, different memory — same tool.

Try this: Identify the AI workflow where you paste the same instructions every time. Create a Project for it — add your custom instructions, upload the files you always reference, and set the rules once. Next conversation in that workspace, notice how much setup time disappears.

Layer 2: Orchestration (The Execution Layer)

This is where work actually gets done. Three blocks that range from simple instructions to autonomous execution.

5. Prompt

Everyone knows this one. Natural language instructions you provide during a conversation — ephemeral, conversational, reactive.

“Review this customer email and suggest three response options: 1) Apologetic and immediate, 2) Professional and measured, 3) Empowered with alternatives.” That’s a prompt. Close the chat and it’s gone. For a one-off request, it’s the right tool.

The mistake people make isn’t at the prompt level — it’s staying there too long. If you find yourself writing the same prompt for the third time, you’ve outgrown this building block.

Try this: Start saving your best prompts as markdown files. The ones that produce great results — save them, refine them, reuse them. This simple habit is the first step toward Skills, which is exactly where we’re headed next.

6. Skill

This is the building block I’m most passionate about, because it’s where most people get the biggest immediate ROI.

A Skill is a reusable instruction package that an AI discovers and loads dynamically when relevant. Think of skills as training manuals, not employees — they teach the AI how to do something without executing autonomously. They’re portable, composable, and loaded progressively.

Here’s what that looks like: you say “Research our competitor Acme and write a competitive brief aligned with my brand guidelines.” The AI discovers the relevant skills on its own — researching-competitors, writing-competitive-brief, applying-brand-guidelines — and loads them. You didn’t tell it which skills to use. It figured that out from the request.

I run my entire content operation this way. I have 68 skills that handle everything from writing LinkedIn posts to preparing cohort sessions. Each one encodes the best practices, templates, and quality standards I’d otherwise have to re-explain every time. Write it once, reuse it everywhere.

Try this: Next time you’re about to work on something you do repeatedly — a weekly report, a client proposal, a content brief — pause and ask AI to help you codify it. “Help me turn this into a repeatable how-to manual with steps, quality criteria, and examples of good output.” That manual is your first Skill. Every time you do that task going forward, it gets faster and more consistent.

7. Agent

Agents are where the hype lives — and where the most misunderstanding happens.

An agent is a system where an LLM controls workflow execution to achieve a goal. It plans, uses tools, reflects on progress, and operates with varying autonomy: semi-autonomous (proposes steps, you approve), autonomous (executes fully, reports when done), or continuous (runs on a schedule without being asked).

The best way to understand agents is to contrast them with skills using the same task:

With a Skill: “Write a LinkedIn post about AI building blocks.” The skill loads your brand voice, post structure, and engagement hooks. It produces the post. You review and publish. The skill didn’t decide to write a post — you did.

With an Agent: “Promote this week’s newsletter across my social channels.” The agent reads the newsletter, decides which platforms to target, identifies the key angles for each, invokes the LinkedIn writing skill for LinkedIn, the X writing skill for X, and drafts all the posts. The agent decided what to do and orchestrated the skills to do it.

Skills teach HOW. Agents EXECUTE. Together they’re the core of the orchestration layer.

I’ll be direct about something: more autonomy isn’t always better. I’ve watched teams spend weeks building autonomous agents for tasks that a simple deterministic workflow would handle in an afternoon. Match the autonomy level to the complexity of the task, not to what sounds most impressive in a demo.

Try this: Create a simple agent as a markdown file. Define its role, the skills it can use, and the task it handles — something like a “Weekly Report Agent” that gathers data, applies your formatting standards, and drafts the report. On platforms like Claude, these markdown-based sub-agents can perform actions, invoke skills, and even coordinate with other agents. No code required — just a well-structured file.

Layer 3: Integration (The Connection Layer)

The first seven blocks can all live inside a chat window. These last three are how AI breaks out of that window and operates inside your actual business systems. For many leaders, this is where the ROI conversation gets real.

8. MCP (Model Context Protocol)

Think of MCP as “USB for AI” — an open standard that lets AI assistants read from and write to your external systems.

Before MCP, AI lived in a chat bubble. You’d copy data from Notion, paste it into the chat, get a result, then copy it back. MCP eliminates that friction. Now AI can check your calendar, update your CRM, post to Slack, and query your database — all within a single conversation.

MCP connections come in three flavors: data sources (read from databases, docs, wikis), action tools (create tasks, send emails, update records), and real-time systems (live data feeds, webhooks).

Here’s a nuance people miss: MCP and Skills serve different purposes. Skills teach the AI what to do. MCP gives the AI access to do it. A “Schedule Team Meeting” agent needs MCP connections to Google Calendar, Slack, and Notion — plus a Skill that teaches it your meeting scheduling preferences. The blocks are designed to work together.

Try this: List the 3 external systems you interact with most during AI workflows. Anywhere you’re manually copying data in and out is an MCP candidate.

9. API

The programmatic interface for accessing AI models from code — stateless, programmable, usage-based. Every AI product you use — from ChatGPT to your favorite writing tool — is built on an API. When you outgrow the chat window and need AI as a service inside your own systems, this is the building block.

A concrete example: a script calls an AI API (Claude, OpenAI, Gemini) to classify 1,000 customer support tickets overnight. Each ticket gets a category, priority, and suggested response. No human in the loop. Cost: a few dollars for work that would take a team member two full days.

Here’s what’s changed recently: you no longer need to be a developer to use APIs. Agentic coding tools like Claude Code and OpenAI Codex let you describe what you want in plain language, and the AI writes and runs the code for you. The API is the building block. The coding tool is how non-developers access it.

Try this: Think of a system you need to access programmatically but don’t know how to call the API — maybe pulling data from your CRM, syncing records between tools, or automating a report. Open an agentic coding tool like Cursor, Claude Code, or OpenAI Codex and describe what you need: “Write a script that pulls all deals closed this month from HubSpot and formats them as a CSV.” The coding tool builds the API integration for you. You don’t need to know the API — you just need to know what you want.

10. SDK

SDKs sit on top of APIs and add the orchestration plumbing — agent loops, tool calling, error recovery, multi-agent coordination. If the API is a raw interface (send request, get response), the SDK is a framework that handles the patterns you’d otherwise build yourself.

Example: using an agent SDK (Anthropic’s Agent SDK, OpenAI’s Agents SDK, or Google’s ADK) to build a research agent that searches the web, reads PDFs, cross-references sources, and produces a structured report. The SDK handles orchestration, iteration, and error recovery. You focus on defining outcomes.

Same story as APIs — you don’t need to be a developer. The same agentic coding tools that unlock APIs also unlock SDKs. Describe the agent you want to build — its goal, tools, and workflow — and let Claude Code or Codex scaffold it. You define the strategy; the coding tool handles the implementation.

Try this: Open an agentic coding tool and describe a simple agent: “Build me an agent that reads a CSV of customer feedback, categorizes each entry, and writes a summary report.” Watch how fast the SDK goes from concept to working prototype.

How Building Blocks Compose

You don’t use all 10 blocks. You compose the ones your workflow needs — and knowing the right combination is the real skill.

Prompt + Context — The starter combo. You provide instructions and the knowledge AI needs. Handles most one-off tasks: writing, analysis, summarization. This is where 80% of people stop, and for many tasks, it’s enough.

Prompt + Context + Skill — Add repeatability. Instead of crafting the same long prompt every time, the Skill encodes your best practices. Same quality, every time, without the setup tax.

Skill + Agent — The orchestration duo. The agent decides what to do; skills teach it how. An agent writing a newsletter might invoke a research skill, a brand voice skill, and a formatting skill — all without you specifying which ones.

Skill + Agent + MCP — Add real-world connectivity. The agent can now read from and write to external systems. This is where AI workflows start replacing manual processes end-to-end.

Where People Get Tripped Up

I’ll share the patterns I see repeatedly in my courses and consulting work, because they’ll save you time.

The biggest one: treating agents as the default. More autonomy adds complexity, latency, and cost. A deterministic workflow for predictable tasks is faster, cheaper, and more reliable. I’ve watched teams spend six figures building an autonomous agent when a 20-line script would have done the job. Match autonomy to the task, not to the hype.

The second: confusing Memory and Context. This trips up nearly everyone. Context is knowledge you provide. Memory is knowledge the AI accumulates from working with you. They solve different problems, and mixing them up leads to workflows that re-upload documents the AI should already know — or expect the AI to remember files you never shared.

Third: dismissing Skills as “fancy prompts.” They’re not. Skills are portable, reusable instruction packages with templates, examples, and quality criteria. Prompts are ephemeral. The difference is the difference between a recipe card and a cooking class.

And finally: assuming you need all 10 blocks. Most workflows need 2-4. Starting with Prompt + Context and adding blocks as complexity demands is almost always the right approach. Over-engineering a simple task is worse than under-engineering a complex one.

Your Action Plan

You don’t need to master all 10 blocks. Start where you are:

If you’re using AI for one-off tasks — You’re at Prompt + Context. That’s a solid foundation. Focus on improving your context quality (better files, clearer instructions) before adding more blocks.

If you’re repeating the same prompts — You’re ready for Skills. Convert your best prompts into reusable instruction packages. This single step eliminates the most common source of inconsistent AI output.

If you’re managing multi-step workflows — You’re ready for Agents. Start with semi-autonomous (agent proposes, you approve) before moving to fully autonomous. Build trust incrementally.

If you’re manually copying data between AI and tools — You’re ready for MCP. Connect your most-used systems and watch the copy-paste tax disappear.

If you’re processing data at scale — You’re ready for APIs and SDKs. Move from chat-based interaction to programmatic access.

The framework isn’t a ladder you climb from bottom to top. It’s a toolkit you draw from based on the job in front of you.

Explore each block in detail. The Hands-on AI Cookbook has a dedicated Agentic AI Building Blocks section with guides, comparison tables, and a “Which block should I use?” decision guide for each of the ten blocks.

The building blocks give you the vocabulary. But the real leverage comes from your judgment — knowing which blocks to compose, when to add autonomy, and when to keep it simple. AI amplifies the clarity you bring to it.

Stay curious. Stay hands-on.

James

Ready to build, not just read? In my Maven courses, we don’t just learn the building blocks — we build with them. Students walk away with AI-powered workflows that accelerate business value.

TL;DR:

• Most people skip from “I want AI to do X” straight to the tool — and wonder why the results are mediocre.

• The missing step is decomposition: breaking your workflow into discrete steps before you build anything.

• The Business-First AI Framework (Discover → Deconstruct → Build) is a three-step process that closes this gap.

• Deconstruct is the step everyone skips — and it’s the most important one.

• Everything is open source. Try it right now — links to get started below.

I watched a student spend two hours building something impressive, only for it to completely break.

She’d taken about ten Google Slides — detailed internal procedures her team uses for pipeline analysis — and packed them into a Google Gem. The setup was sophisticated. She’d included scoring rubrics, qualification criteria, and stage-by-stage review processes. Everything her team’s top performers actually do when they evaluate a deal.

The Gem couldn’t deliver.

Not because the AI wasn’t capable. Not because Google Gems are limited. The instructions she’d written described what the pipeline review should accomplish — but not how her team actually does it at the step-by-step level. The scoring rubrics were there, but the decision logic connecting them wasn’t. The criteria were listed, but the sequence for evaluating them — which signals what to check first, when to escalate, and what constitutes a red flag versus a yellow flag — lived in her head, not in the instructions.

When she realized this, you could see the shift. “The AI can’t read my mind,” she said. “I need to describe this the way I’d train a new analyst — not the way I’d summarize it for my boss.”

That’s the aha moment. And I see some version of it every single cohort.

The pattern is always the same. Someone has a clear idea of what they want AI to do. They go straight to the tool. They spend an hour — sometimes several — iterating on prompts, rearranging context, trying different approaches. The output is decent but never quite right. Eventually, they either settle for “good enough” or give up.

The problem isn’t the tool. It’s the step they skipped.

The Idea-to-Implementation Chasm

Most AI education teaches you tools. How to write better prompts. Which model to use for which task? How to set up a GPT, a Gem, or a Project.

Almost nobody teaches the bridge: how to go from “I want AI to do X” to a working, repeatable workflow that actually performs.

This matters because people are trying to reach three very different outcomes with AI — and they often don’t realize which one they need:

1. Collaborative AI — You drive the process, AI contributes. Co-writing a document, brainstorming a strategy, and reviewing code together. The AI is a capable partner, but you’re steering.

2. Deterministic Automation — A repeatable process that AI executes reliably with minimal supervision. Formatting reports, processing forms, and generating standardized communications. You define the rules once; AI follows them every time.

3. Autonomous Agents — AI plans and executes multi-step work independently. Research pipelines, monitoring systems, and multi-stage content production. You set the goal; AI figures out the steps.

Without a structured approach to get there, most people default to option 1 — chatting back and forth — even when what they actually need is option 2 or 3. They’re using AI as a conversation partner when they should be using it as an engine.

The Business-First AI Framework

I built this framework because I kept watching the same failure mode play out.

People would come to my cohorts with real workflows they wanted to automate. Good ideas, real pain points. But they had no standardized way to articulate what the workflow does and why each step matters — at the level of detail AI needs to execute it. There was no business methodology for translating “here’s what my team does” into a specification that an AI could work from.

So they’d skip straight to the how. Open ChatGPT, start writing prompts, iterate for hours. The problem isn’t that they lacked technical skill. The problem is they were trying to build before they’d defined what they were building. It’s like writing code before you’ve written requirements — you’ll produce something, but it won’t be what you actually need.

The Business-First AI Framework puts the what and the why first. Three steps, and the order matters.

Step 1 — Discover.

Audit your workflows and identify which ones are candidates for AI.

Before you can apply AI to anything, you need to know where it fits. Step 1 is a structured audit that scans your role, responsibilities, recurring tasks, pain points, and multi-step processes — then produces a prioritized list of opportunities categorized as Collaborative AI, Deterministic Automation, or Autonomous Agent.

The key insight: don’t start with the technology and ask “where should we use it?” Start with your workflows and ask, “Which of these would benefit most from AI?”

Most people discover 5-15 opportunities they’d never considered. The audit surfaces patterns you miss in the daily grind.

Step 2 — Deconstruct.

Break the chosen workflow into atomic steps.

This is the step everyone skips. And it’s the most important one.

You describe your workflow — rough and incomplete is fine — and then systematically decompose each step using five questions:

1. Discrete steps — Is this actually one step, or multiple steps bundled together?

2. Decision points — Are there if/then branches, quality gates, or judgment calls?

3. Data flows — What goes in? What comes out? Where from and where to?

4. Context needs — What documents, files, or reference materials does this step require?

5. Failure modes — What happens when this step fails?

A workflow that starts as 5-8 rough steps typically expands to 12-20 refined steps after this process. That expansion is the point — it surfaces every hidden sub-step, every implicit decision, every assumption you’d internalized so deeply you forgot it was there.

Remember the Google Gem student? Had she run her pipeline review process through these five questions first, she would have seen immediately that her instructions lacked decision logic. The scoring rubrics were there, but the workflow connecting them — the sequence, the branching, the escalation criteria — wasn’t described at the level of detail the AI needed to execute it.

The deliverable from Step 2 is a Workflow Definition—a structured Markdown file that captures every step in detail. It’s platform-agnostic. It works whether you’re building for ChatGPT, Claude, Gemini, or anything else.

Step 3 — Build.

Design the AI implementation, then construct it.

Step 3 takes your Workflow Definition and turns it into a working AI workflow through three parts:

• Design — Choose your execution pattern. A simple prompt? A skill-powered prompt with reusable routines? A single autonomous agent? A multi-agent pipeline? The right choice depends on what your workflow actually needs — not on what sounds most impressive.

• Construct — Build only what your pattern requires. Map each step to the right AI building block: prompts, context documents, skills, agents, or tool connections. The Building Block Spec from Design tells you exactly what to create and in what order.

• Run — Execute the workflow on a real scenario. Evaluate. Iterate. Most workflows require 2-4 rounds of refinement before producing reliably good output.

The output is a Baseline Workflow Prompt you can paste into any AI tool and run immediately. For more complex workflows, you get agent configurations and skill definitions that you can deploy to Claude or adapt for other platforms.

Why Deconstruct Changes Everything

Two patterns I see repeatedly:

The “I described the what, not the how” pattern. Like the Google Gem student. People describe the outcome they want — “review my pipeline deals and score them” — without describing the process at the level of detail AI needs to actually do it. Deconstruction forces you to articulate every step, which reveals what was implicit.

The “skip to the tool” pattern. People start iterating on prompts immediately. They spend hours refining wording, when the real problem is that they never defined what “done” means at each step. Deconstruction front-loads that thinking, so the build phase goes faster.

When people see the five-question framework applied to their own workflow, two things click:

First, they understand how to go about it. The five questions are a methodology—not a vague “think harder about your process” — but a structured interview that systematically surfaces detail. It replaces guessing with a repeatable approach.

Second, they see the level of detail required for AI to orchestrate something specific to their business. Not generic instructions anyone could write, but the particular decision logic, context documents, quality criteria, and failure handling that make a workflow theirs.

This is the moment the framework earns trust. Not when I explain it — when they use it on their own work and see the difference in output quality.

Try It Right Now

You can start Step 1 — Discover — right now. Go to the Discover Workflows page on the Hands-on AI Cookbook:

handsonai.info/business-first-ai-framework/discover

The page gives you two options:

• Any AI tool — Copy the prompt template and paste it into ChatGPT, Claude, Gemini, M365 Copilot, or whatever you use. It runs a structured audit of your workflows and produces a categorized report of opportunities.

• Claude platform — If you use Claude Code, Cowork, or Claude.ai, install the Business-First AI plugin (/plugin install business-first-ai@handsonai) and the skill runs interactively with file-based deliverables.

Either way, you’ll walk through a guided conversation that takes about 20 minutes. Most people discover 5-15 opportunities they’d never considered. Pick the one that excites you most — that’s your candidate for Step 2.

Already know what you want to build? Skip straight to Step 2 — the step everyone misses:

handsonai.info/business-first-ai-framework/deconstruct

Describe your workflow — rough and incomplete is fine — and the prompt walks you through the five-question deep dive. In 15-25 minutes, you’ll have a Workflow Definition that captures every step, decision point, and failure mode. This is the part that transforms “I want AI to do X” into something AI can actually execute.

What the output actually looks like

To make this concrete, I ran my own Content Calendar Planning workflow through the framework — the process I use every Sunday to plan two weeks of content across LinkedIn, Substack, X, and YouTube.

“I plan content on Sundays” became 10 refined steps across four phases, with decision points, data flows, failure modes, and a dependency map. The framework then produced an AI Building Block Spec that classified each step on the autonomy spectrum, identified four reusable skills, and recommended a build order — starting with the conversational planning flow (no infrastructure needed) and layering in database skills incrementally. Finally, it generated a ready-to-run Baseline Workflow Prompt that orchestrates the entire 10-step process as a collaborative conversation.

See the full example with all three deliverables:

Content Calendar Planning — Full Worked Example

The page shows the complete Workflow Definition, AI Building Block Spec, and Baseline Workflow Prompt — everything the framework generated for a single real workflow, from decomposition through to a working prompt you could paste into any AI tool.

Three Ways to Use the Cookbook

Everything I’ve described — the framework, the prompts, the worked examples — lives in the Hands-on AI Cookbook, an open-source repository anyone can use. Here’s how to get started, depending on how you work:

1. Browse and copy. Visit handsonai.info in your browser. Every framework step has a ready-to-use prompt template. Copy it, paste it into ChatGPT, Claude, Gemini, M365 Copilot — whatever you use. No account required. No vendor lock-in.

2. Connect via MCP (recommended). If your AI tool supports MCP (Model Context Protocol), you can bring the entire cookbook knowledge base directly into your conversations. Add this URL as a connector and your AI assistant can search, read, and reference any cookbook page on your behalf — framework guides, building block definitions, worked examples, everything:

MCP server URL: https://mcp.handsonai.info/mcp

This works in Claude, ChatGPT, Cursor, VS Code, and any MCP-compatible tool. Setup takes about 30 seconds. Once connected, just ask your AI: “Search the Hands-on AI Cookbook for how to deconstruct a workflow” — and it pulls the relevant guide directly into your conversation. No tab-switching, no copy-pasting.

Setup instructions for every supported tool

3. Install the plugin (Claude users). For the full interactive experience in Claude Code or Cowork, install the Business-First AI plugin:

/plugin install business-first-ai@handsonai

The plugin implements all three framework steps as executable skills. Describe what you need — “Help me deconstruct my weekly reporting process” — and the framework runs interactively, saving deliverables as structured files you can iterate on.

Everything is plain-text Markdown. No compiled code, no black boxes. Read the source, understand how it works, and adapt it for your own workflows.

Your Turn

You have a workflow right now that’s costing you hours. You know which one it is — you thought of it while reading this.

Don’t open ChatGPT and start prompting. That’s the trap.

Go to the Discover page and run the audit. Or if you already know the workflow, go straight to Deconstruct and break it down. Twenty minutes of structured decomposition will save you hours of aimless iteration.

What workflow are you going to deconstruct first? Reply and tell me. I read every response and use your feedback to improve the cookbook — your sticking points become new guides, better prompts, and worked examples that help everyone.

Know someone stuck in the idea-to-tool gap? Forward this to them.

— James

P.S. If you want structured guidance through all three steps with a cohort of peers, I teach two live courses: Hands-on Agentic AI for Leaders — for leaders and professionals applying AI to their business workflows using popular tools like ChatGPT, Claude, Gemini, and M365 — and Claude and Claude Code for Builders — for people building workflows and apps on the Claude platform. Every student walks out with at least one fully deconstructed, working AI workflow. See upcoming cohorts at courses.handsonai.info

TL;DR:

• Most people are stuck between “I’ve tried ChatGPT” and “I’m getting real results.” The gap isn’t tools — it’s a system.

• I built the Hands-on AI Cookbook (handsonai.info) — a free, open-source resource for turning AI ideas into production.

• It covers a 3-step framework, six use cases where AI creates value, six building blocks that work across Claude, ChatGPT, Gemini, and M365 Copilot, and ready-made tools you can install with a single command.

• No coding required for the strategy content. Builder content scales from foundational to advanced.

• It’s free, open-source, and always evolving. Visit handsonai.info.

If you have a business workflow you want to make AI-first, the last thing you need is another slick demo. You need clarity — what are the building blocks, how do they fit together, and how do you go from idea to production?

That’s what the Hands-on AI Cookbook is. Complex AI topics distilled into clear frameworks, building blocks, and practical tools so builders can stop experimenting and start operationalizing.

And by builders, I don’t mean developers. If you’re a leader, entrepreneur, or professional with a business idea and the willingness to get hands-on, you’re a builder.

Why I Built This

After working with 6,000+ executives at Berkeley Haas through enterprise AI programs and hands-on cohorts, I kept seeing the same pattern. People would show up excited about AI. They’d tried the tools. They’d written some prompts. But they couldn’t close the distance between experimenting and executing.

The gap wasn’t about prompts. It was about thinking. People didn’t need another “50 prompts for marketers” listicle. They needed a framework for figuring out where AI fits in their work and how to build it into their workflows — systematically, not randomly.

So I started building. Openly, publicly, on GitHub. Because I believe AI knowledge should be free and open to everyone.

Here’s the thing — we all have a finite amount of time. The best gift I can give anyone is time and clarity to maximize their impact. My hope is that this cookbook shortens the distance between where you are and what you’re capable of — bringing the future forward, so you can spend your time on the work that actually matters.

The result is handsonai.info — a living, open-source cookbook with everything I’ve learned about applied AI, organized so you can actually use it.

What’s Inside

The cookbook is built in four layers. Each one takes you deeper.

Layer 1: A Framework for Going from Idea to Production

Most people have a sense of which workflows could benefit from AI. The hard part is going from that idea to something that actually runs — reliably, repeatedly, in the real world.

The Business-First AI Framework guides you through that entire journey:

1. Discover — Run a structured audit of your workflows to identify where AI creates the most value

2. Deconstruct — Break those workflows into discrete steps and map each one to AI building blocks

3. Build — Turn that analysis into a working AI workflow, ready for production

You don’t need to be technical to use any of this. The framework works on paper, in a conversation with any AI tool, or with installable plugins that walk you through it step by step. The Discover step takes 20 minutes and produces a prioritized list of AI opportunities — a tangible output in one sitting.

Layer 2: Use Cases — Where AI Creates Value

The cookbook includes six use case primitives — Content Creation, Research, Coding, Data Analysis, Ideation & Strategy, and Automation — adapted from OpenAI’s analysis of 600+ enterprise AI deployments. These primitives describe what type of work AI actually does in the real world. They give you a classification system so you can look at any workflow and immediately know which category it falls into and which building blocks to reach for.

Layer 3: The Building Blocks

Every AI workflow — from a single prompt to a multi-agent pipeline — is assembled from six building blocks:

• Prompt — Instructions you give the AI

• Context — Background knowledge the AI needs (your data, your docs, your domain)

• Project — A persistent workspace that holds everything together

• Skill — A reusable routine the AI can invoke on demand

• Agent — An autonomous AI that plans and executes multi-step work

• MCP — A connector that lets AI access external tools and data

These are platform-agnostic concepts. The cookbook maps each block across Claude, ChatGPT, Gemini, and M365 Copilot in a single comparison table. So instead of learning four different systems, you learn the six blocks once and apply them everywhere.

This is the vocabulary that makes AI adoption repeatable. Once you understand building blocks, you can look at any workflow and say: “This needs a skill with context and an MCP connection” — regardless of which platform you’re using.

Layer 4: Accelerators — Domain Expertise You Can Install

As I learn, I codify that knowledge into AI agents and skills — pre-built tools you can install in one command to accelerate your own learning and implementation. This is an evolving library, not a finished product.

The cookbook’s Plugin Marketplace is where these live. The Business-First AI plugin alone packages 8 agents, 6 skills, and 3 prompts that implement the entire Business-First AI Framework as executable tools. Tell it, “help me find where AI can improve my workflows,” and it runs a structured audit. Tell it “deconstruct my client onboarding process,” and it walks you through the full analysis.

There’s also a Builder Stack Setup Guide. This is where I see people get stuck — the tools for building and running real AI workflows are unfamiliar to most people. Terminal, Git, AI coding CLIs — many of these will be new. But they’re what unlock the ability to go from idea to production. The guide walks you through 7 steps, roughly 75 minutes, with checkboxes to track your progress. One VP of Product used it to go from zero terminal experience to a fully configured AI builder stack in a single afternoon.

And the cookbook is full of patterns, walkthroughs, and direct answers to the questions that come up most often — from “what is a system prompt?” to “how do I schedule an AI agent to run automatically?”

Who It’s For

Anyone stuck between “experimenting” and “executing” — leaders auditing where AI fits, entrepreneurs building their first AI-powered workflow, professionals who want to get hands-on and take an idea all the way to production.

No coding experience required for the strategy content. The builder content scales from first-time terminal users to teams orchestrating multi-agent pipelines.

Go Build Something

In a world changing this fast, there are no shortcuts. You have to get hands-on with the tools. You have to touch them, experiment, and develop your own sense of what’s possible.

I’ve trained CEOs and C-suite leaders on the same tools covered in this cookbook. Every one of them came out with greater clarity — not because AI gave them answers, but because getting hands-on calibrated their mindset to what’s actually possible.

That’s my challenge to you: start small. Build the muscle of a business-first approach to AI. Value creation isn’t just the power of the tools — it’s your ability to innovate from within and evolve what you’re capable of. That inner transformation is what lets you, together with AI, bring out your full potential to make your ideas come to life.

The Hands-on AI Cookbook is free, open-source, and always evolving.

Start here: Visit handsonai.info and run the Discover step. In 20 minutes, you’ll have a prioritized list of AI opportunities for your team — ranked by impact and ready to act on.

If it helps you, share it with someone who’s stuck between experimenting and executing. That’s exactly who it’s built for.

Stay curious and hands-on,

James

P.S. Know someone who keeps saying “I need to figure out this AI stuff”? Forward them this post. The cookbook is the fastest way to go from curious to capable.

A leader wants to automate prospect research. Or customer outreach. Or content creation. They’ve done the process hundreds of times. They know it works. But when they try to apply AI?

They get stuck.

Not because the AI isn’t capable. But because they’ve never had to explain their process with the kind of precision an AI needs. The workflow lives in their head as intuition—not as clear, repeatable steps.

On Friday, I hosted a Lightning Lesson where we tackled this together. Over 340 leaders and professionals joined me to walk through how to take something you know intimately and break it down so AI can actually execute it.

No theory. Just a real workflow, deconstructed step by step.

The Challenge: Getting What’s in Your Head Into Structure

Think about something you do regularly in your business. Maybe it’s qualifying leads. Analyzing feedback. Preparing reports.

Now imagine explaining every single step to someone who’s smart but has never done it before. Not just the what—the how, the why, the decision points, the nuances.

Suddenly it’s not so simple, right?

That’s the gap. You have expertise that’s second nature. AI needs explicit instructions. Deconstruction is how you bridge that gap.

And honestly? The process of breaking it down often reveals things you didn’t even realize you were doing—which makes your process better, with or without AI.

What We Built Together

In this session, I walked through a workflow I use for LinkedIn prospect research. Nothing fancy—just a practical business process:

• Start with a buyer persona (the kind of prospect you want to find)

• Search LinkedIn for people who match that profile

• Evaluate each prospect against specific criteria

• Generate personalized engagement recommendations

• Output everything in a structured format

The interesting part isn’t the workflow itself. It’s how we approach building it.

Key Concepts You Can Apply Immediately

1. The “What/Why vs. How” Principle

You shouldn’t be writing detailed AI instructions from scratch. That’s the old way.

Instead, you define the business outcome and sketch high-level steps. Then let AI generate the detailed execution instructions.

You bring domain expertise. AI brings execution precision. Stay in your lane.

2. Meta-Prompting: Let AI Write AI Instructions

Here’s the actual technique:

“You are an expert workflow designer and prompt engineer. Please write a prompt for this scenario. The outcome is [your goal]. Here are the high-level steps: [your steps]. Now write the detailed instructions.”

Let the model craft the “how” while you focus on the strategic “what and why.”

I demonstrate this live in the session—you’ll see how much better the AI-generated instructions are than what most people write manually.

3. Skills vs. MCPs: Understanding the Difference

This came up multiple times in Q&A because it’s genuinely confusing.

Skills teach Claude how to do something. Procedural knowledge. “Here’s how to write a LinkedIn post in my style.”

MCPs (Model Context Protocol) give Claude access to something. Tool connectivity. “Here’s how to read from and write to my Notion database.”

They work together. Skills provide the methodology. MCPs provide the capability.

4. Build a Workflow Registry

Don’t just build workflows ad hoc. Create a system.

I show my Notion setup where every workflow is documented with:

• Name and business process assignment

• Description and expected outcome

• Trigger conditions

• The actual steps

• Links to AI assets (prompts, personas, templates)

• Status tracking

This becomes your institutional knowledge. Your competitive moat.

5. The Clarity Test

Here’s how you know if a workflow is ready to automate:

Can you explain it clearly enough that a smart person who’s never done it could execute it successfully?

If not, you’re not ready for AI yet. The work is in the deconstruction, not the automation.

6. Create Reusable AI Assets

In the demo, I use a buyer persona stored as a markdown file. It’s an AI asset I can plug into multiple workflows—prospect research, email outreach, and content creation.

Think in building blocks. What pieces of knowledge or context can you document once and reuse everywhere?

Questions That Came Up

The Q&A was where things got practical. People asked questions like:

“Should I design my Notion database first, or let Claude do it?” (Start simple with what makes sense to you, then let Claude optimize based on your actual workflow)

“When do I need a Skill versus just a good prompt?” (Skills when you’re doing the same thing across multiple workflows and want consistent execution)

“How detailed should my workflow steps be?” (Detailed enough that the AI knows what to do, but not so prescriptive that you lose flexibility)

These weren’t hypothetical. These were people actively working through this in their businesses, hitting real obstacles, finding real solutions.

Why This Matters Right Now

We’re past the “playing around with ChatGPT” phase. The tools are ready. The question isn’t whether AI can help your business—it’s whether you can articulate your processes clearly enough to take advantage of it.

The leaders who figure this out aren’t necessarily the most technical. They’re the ones who can think operationally, break down their expertise, and build systems that scale.

That’s what this Lightning Lesson is about. Not the technical wizardry (though we cover that too). The mindset shift from “What can AI do?” to “What do I need done, and how do I break it down?”

Want to Go Deeper?

This Lightning Lesson gives you the approach. If you want to actually build these systems with hands-on guidance and expert feedback, I’m running two cohort courses:

Claude and Claude Code for Builders – Starts Tomorrow (January 26)

This course is primarily for “builders” - business people who want to go deep on Claude’s capabilities, Claude Code for agentic workflows, and building a prototype application (e.g., website)

25% founder discount for this inaugural cohort.

View syllabus and enroll →

Hands-on Agentic AI for Leaders – Next cohort starts February 2

This is for business leaders and non-technical builders who want to move from experimentation to actually deploying AI in their operations. We build real workflows, deploy them, and develop the literacy to lead AI transformation.

Rated 4.8/5. Over 250 students trained.

View syllabus and enroll →

The best AI implementation starts with clear thinking about your business, not with fancy prompts.

Watch the session. Pick one workflow. Break it down.

That’s where real progress starts.

— James

Subscribe now

Daily Nugget

“The easiest way for us to gain happiness is to learn how to want the things we already have.” - A Guide to the Good Life by William B. Irvine

Hey, everybody. It’s Monday, January 19th, and I want to share something with you.

I love books. Every day, I read from The Daily Stoic by Ryan Holiday and The Daily Laws by Robert Greene. A lot of times, we need reminders of really important things—basic things, even—but we need to hear them again. When I find nuggets, I want to share them with you.

One book that helped me through a difficult period years ago is The Guide to the Good Life. It’s based on Stoic philosophy, and there’s a chapter in here that I keep coming back to. It’s about hedonic adaptation—this very human tendency to be insatiable.

The Satisfaction Treadmill

Here’s what the author says:

“We humans are unhappy in large part because we are insatiable.”

You know this treadmill. We achieve something we’ve worked hard for, and almost immediately, we want more. There’s nothing wrong with being successful or wanting to grow. But it’s when that pursuit starts controlling our lives and making us unhappy that we need to pause.

The author explains it this way:

“We are unhappy when we detect an unfulfilled desire in ourselves. We work hard to fulfill this desire in the belief that on fulfilling it, we will gain happiness. The problem, though, is that once we fulfill a desire for something, we adapt to its presence in our life, and as a result, we stop desiring it—or at any rate, we don’t find it as desirable as we once did. We end up just as dissatisfied as they were before fulfilling the desire.”

Your job. Your relationship. Your home. The things we once dreamed of having, we now take for granted.

The Solution

So what’s the answer?

The author writes:

“One key to happiness is to forestall this adaptation process. We need to take steps to prevent ourselves from taking for granted, once we get them, the things we worked so hard to get.”

And here’s the nugget—I have this highlighted because it’s so true:

“The easiest way for us to gain happiness is to learn to want the things we already have.”

This advice is easy to state. The trick is putting it into practice. How do we convince ourselves to want the things we already have?

Your Assignment Today

I talk to a lot of people who are bitching and complaining about things that, frankly, are irrelevant. I think if we really embrace this idea of loving the things we already have, we’ll not only be happier—we’ll probably be less stressed as we go throughout our days.

So here’s my challenge:

Take a pause and think about all the amazing things you have in your life. Your health. Your family. Your friends. Really embrace those and love them.

And ask yourself: Is there something you’re pursuing that’s driving you astray because you feel unsatisfied? Because you’re insatiable for that thing?

This was the nugget I reminded myself of today. I’m hoping this short reflection can give you a dose of happiness, too.

What’s one thing you already have that you could appreciate more today? Hit reply—I’d love to hear from you.

→ Your workflow instructions are probably too vague or too long (300-line mega-prompts don’t work)
→ Claude Skills package your expertise into reusable how-to manuals that activate automatically
→ The process: Map workflow → Break down steps → Build skills → Test and reuse
→ Live demo: Building a meeting notes organizer skill in real-time
→ One simple prompt can trigger multiple skills that execute complex workflows flawlessly

Subscribe now

I just wrapped my Lightning Lesson on “Design Your First Skill-Powered Workflow” and wanted to share the recording with you.

Most people get workflow instructions wrong in one of two ways:

Too high-level: “Research our competitor and write a brief.” Claude guesses. You get mediocre results.

Too detailed: A 300-line mega-prompt that hits context limits and still doesn’t perform.

The real issue? Tasks that require precise procedural control—the kind you execute manually with rigor—need a different approach.

That’s where Claude Skills come in.

What You’ll Learn in This Session

The mental model: Think of Skills as instruction manuals. When you ask Claude to “organize meeting notes,” it automatically reads the how-to manual you’ve created, complete with edge cases, examples, and resources.

The activation pattern: Claude reads all your skill metadata when you boot up. When keywords in your prompt match a skill, it pulls the full instructions and executes. No more explaining the same process 50 times.

The build process:

1. Map your workflow (business process → workflows → steps)

2. Identify skill-worthy tasks (repeatable + need rigor)

3. Collaborate with Claude to build the skill

4. Test it and add to your toolbox

Live demonstration: I built a meeting notes organizer skill from scratch in the session—you’ll see the exact questions Claude asks, how to answer them, and what the final skill looks like.

Real workflow example: I showed my Lightning Lesson creation workflow that went from this 300-line manual process to a single prompt:

“Create a new Lightning Lesson on [topic]”

→ Claude designs the lesson, creates a Word doc, and saves it to the Notion database. Three skills activated automatically.

Why This Changes Everything

Here’s what clicked for attendees:

Skills aren’t just about saving keystrokes. They’re about packaging your domain expertise so Claude doesn’t have to read your mind.

When you build a skill, you’re creating IP. Reusable assets you own. A war chest of procedures that compound over time.

I’ve built 50+ skills for my one-person business. Each one unlocks work I used to find too cumbersome to do consistently.

Watch the Full Session

The recording walks through:

• The decomposition framework (Process → Workflow → Skill)

• Live skill creation with real-time Q&A

• How skills activate based on keywords

• My actual Notion database showing skill-powered workflows

• Common mistakes and how to avoid them

Go Deeper: Claude for Builders Course

If you want to get hands-on and go deeper with Claude, Claude Code, and Cowork, join me for a cohort adventure to learn with other builders who want to operationalize high-value use cases.

In 5 weeks, you’ll build:

✅ Foundation: Configure your builder stack and design systematic workflows
✅ Reusable Assets: Build Claude Skills that execute your expertise on demand
✅ Collaborative AI: Deploy workflows where Claude works WITH you
✅ Autonomous Workflows: Build multi-agent systems and browser automations that run independently
✅ Applications: Ship web app prototypes using agentic coding—no engineering required

You get intimate cohorts, 1:1 coaching, and lifetime access.

We build together—not lectures.

First cohort launches Jan 26 — limited to 20 builders.

Use promo code FOUNDER to save 25%, shape the course, and attend again free in 2026.

See the full syllabus →

Your Turn

Drop a comment and tell me: What’s the first workflow you’re going to supercharge with Claude skills?

I read and respond to every comment—and the best ideas might become future Lightning Lessons.

Stay curious,
James

P.S. Two more Lightning Lessons coming up if you want to keep building:

→ Deconstruct Your Workflows for Agentic AI — Friday, Jan 24 (sign up for free ) - Learn a framework to break workflows into AI-executable steps

→ Build Your Agentic Workflow Registry — Friday, Jan 31 (sign up for free)
Map all your processes, workflows, and AI assets in a registry

Lightning Lesson Sequence

Friday, Jan 16: Design Your First Skill-Powered Workflow
Learn to decompose your workflow into steps that reveal which tasks can be encapsulated into reusable skills. Then build and run your first skill-powered workflow.

Sign up for free

Friday, Jan 23: Deconstruct Your Workflows for Agentic AI
Learn to break down your work into steps that AI can execute. This is the foundation—you can’t automate what you haven’t articulated.

Sign up for free

Friday, Jan 30: Build Your Agentic Workflow Registry
Build a system that gives you a holistic view of every workflow you execute and which AI tools operationalize them. You’ll leave with a registry and an approach for managing your IP as a living system.

Sign up for free

Each session is 30 minutes. You can attend all three or just the ones relevant to where you are. No prerequisites—just practical implementation guidance.

If you can’t attend live, still sign up—you’ll get the deck and video right after each event.

If you’re serious about building AI-powered workflows, these sessions will help you structure your approach.

Please share with friends and colleagues who could benefit from these sessions.

See you there!
-James

Look at those highlighted lines.

“The only thing you truly possess is your ability to make choices.”

“This is the only thing that can never be taken from you completely.”

How much time do we waste on things completely outside our control?

The economy. What someone thinks of us. Whether that deal closes. Whether AI disrupts our industry. Whether that person responds to our email.

We ruminate. We strategize. We stress. We lose sleep.

And none of it moves the needle—because we never had the lever to pull in the first place.

The practice I keep coming back to:

A simple pause. Before I spend time or energy on something, one question:

Is this within my control, or outside it?

If it’s outside—I let it go. If it’s inside—I act.

Why it matters:

Time is the only resource we cannot get back. We don’t know how much we have. It’s finite. Every hour spent worrying about what we can’t control is an hour stolen from what we can.

Today—this week—practice the pause. Ask the question.

What’s one thing you’ve been spending energy on that’s actually outside your control?

Sometimes just naming it is enough to let it go.

Good luck this week as you practice this essential skill toward self-mastery.

-James

I Have This Workflow... Now What?

You’ve identified a workflow that’s ready to be reimagined with AI—automated, augmented, or rebuilt from the ground up.

Maybe it’s competitive analysis, content creation, customer research, or course design. You’ve seen what’s possible. You know AI can help. But now you’re stuck with a harder question:

Should I write a better prompt? Create a Project? Build a Skill? Set up an Agent? Connect MCP?

You know this workflow can become AI-first. You’re just not sure which building blocks to reach for.

This decision paralysis is universal. I hear it from executives in my Maven courses, from consulting clients, from readers of this newsletter. Everyone hits the same wall: I know the pieces exist, but I don’t know which ones I need or how they fit together.

The paralysis happens because we don’t understand the building blocks of agentic AI—what each one does, when to use it, and how they work together.

But here’s what I’ve learned after building dozens of workflows and teaching thousands of leaders: the deeper issue isn’t technical. It’s mindset.

Most leaders are still thinking like task executors: “How do I get AI to do this task?”

The shift is to workflow architect: “How do I design a system that accomplishes this goal?”

Each building block represents a different level of architectural thinking:

Prompts = Reactive. You’re in the loop for every action.

Context + Projects = Organized. You’re curating knowledge.

Skills = Systematized. You’re codifying expertise.

Agents + MCP = Orchestrated. You’ve designed a system.

This article will give you the mental model to make confident decisions about which building blocks to use—and more importantly, how to combine them into workflows that actually work.

By the end, I’ll show you exactly how I use all six building blocks together in a real workflow I run every week: designing cohort courses.

Subscribe now

What Makes a Workflow “Agentic”?

Before we dive into building blocks, let’s establish what we’re actually building toward.

Andrew Ng offers the clearest definition: “An agentic workflow is a process where an LLM-based app executes multiple steps to complete a task.”

The key phrase is multiple steps. When you type a question into ChatGPT and get an answer, that’s not agentic—that’s a single inference. When an AI system researches a topic, synthesizes findings, drafts a report, evaluates quality, and revises based on that evaluation, that’s agentic.

But not all agentic systems are created equal. They exist on a spectrum of autonomy:

Deterministic (Less Autonomous) sits at one end. You sequence the steps in advance. You decide which tool to use for each step. The workflow is deterministic and repeatable—same inputs always produce the same outputs. You’re in full control of the process; the AI executes within tightly defined boundaries.

Semi-Autonomous (Semi-Deterministic) occupies the middle. The LLM can make some decisions within each step, but tools are configured in advance. This is where human-in-the-loop workflows live—you maintain oversight and approval authority while the AI handles reasoning and execution. There’s variation in output, but the overall flow is structured.

Fully Autonomous (Agents) sits at the other end. No predefined steps. The agent brain plans the steps. The agent decides what steps to take and calls tools as needed. You set the goal; the agent figures out the route. The path and output vary based on what the agent discovers along the way.

Here’s what most people miss: you don’t always need a fully autonomous agent.

In fact, most business workflows benefit from the middle ground—semi-autonomous workflows where humans stay in the loop for judgment calls. You get the power of AI reasoning without surrendering control over decisions that matter.

Why does the agentic approach outperform simple prompting? Several reasons:

Speed. You can run LLM steps in parallel rather than sequentially.

Higher performance. Models that use extended reasoning can think more deeply and for longer on complex problems.

Modularity. You can use different LLMs optimized for each step—a fast model for simple tasks, a powerful model for complex reasoning.

Repeatability. Deterministic workflows execute consistently every time.

Adaptability. When you don’t know the exact steps required beforehand, agents can plan on the fly.

The building blocks I’m about to walk through are the components you’ll combine to create workflows at whatever level of autonomy your situation requires.

The Six Building Blocks

Let’s examine each building block in detail: what it is, when to use it, and when to reach for something else instead.

1. Prompt

What it is: Instructions you provide to an AI in natural language during a conversation. Prompts are ephemeral, conversational, and reactive—you provide context and direction in the moment.

Three characteristics:

Ephemeral. A prompt exists only within a single conversation. When you start a new chat, you start fresh. There’s no memory, no persistence, no accumulated context.

Conversational. Prompts are written in natural language. You’re not programming; you’re communicating. This makes them accessible but also means they require clarity and specificity to work well.

Reactive. You provide direction in real-time based on what you see. The AI responds, you evaluate, you redirect. It’s a dialogue, not a script.

When to use prompts:

Prompts are your primary interface with AI. Use them for:

• One-off requests. “Summarize this article.” “What are the key themes in this data?”

• Exploring new problems. When you don’t yet know what you need, prompts let you iterate quickly.

• Quick questions or drafts. Low-stakes work where speed matters more than consistency.

• Conversational refinement. “Make that more concise.” “Add an example.” “Adjust the tone.”

• Creative collaboration. Working with AI as a thought partner, bouncing ideas back and forth.

Real example:

When I’m developing a new lesson, I’ll often start with a prompt like: “Review this lesson outline and suggest where students might get stuck. Focus on the hands-on exercise—is it too ambitious for a 90-minute session?”

This isn’t a repeatable workflow. It’s a specific question about a specific artifact in the moment. That’s what prompts are for.

When to use something else:

If you find yourself typing the same prompt repeatedly across multiple conversations, that’s a signal. You’ve identified a pattern worth systematizing. Transform that recurring prompt into a Skill so you don’t have to re-explain the procedure every time.

The shift from prompt to Skill is a shift from reactive to proactive—from explaining what you want in the moment to codifying what you want once so it applies automatically.

2. Context

What it is: Unique knowledge not contained in AI models, required by the workflow for execution. Context is information from sources like documents, databases, files, and proprietary data that the AI wouldn’t otherwise have access to.

Three characteristics:

Uploaded files. PDFs, spreadsheets, images, code files—anything you attach to a conversation or project. The AI can read and reference this information directly.

Referenced documents. Google Docs, Notion pages, Markdown files—documents that exist in external systems that you bring into the AI’s awareness, either by copy-pasting or through connected integrations.

Connected data. Databases, APIs, real-time systems—structured information that the AI can query dynamically. This is where Context intersects with MCP, which we’ll cover later.

When to use context:

Context bridges the gap between what AI knows from training and what it needs to know for your specific situation. Use it when:

• AI needs company-specific information. Your product specs, pricing, org structure, competitive landscape—none of this is in the model’s training data.

• Working with proprietary data. Customer lists, internal metrics, strategic plans—information that should never have been in public training data.

• Referencing brand guidelines or standards. Voice, tone, visual identity, compliance requirements—the rules that govern how work should be done at your organization.

• Grounding responses in facts. When accuracy matters more than creativity, context provides the source of truth.

Real example:

My course curriculum lives in a Notion database: courses, modules, lessons, assignments, and learning objectives. When I’m designing a new course, Claude needs access to this context to understand what I’ve already built, maintain consistency with the existing curriculum, and avoid duplicating content.

Without this context, Claude would be working blind—generating content that might be good in isolation but doesn’t fit my larger system.

When to use something else:

Context provides the what—the information the AI needs to reference. But information alone isn’t instruction. If you need to teach Claude how to process that information—what patterns to follow, which quality standards to apply, and which output format to use—you need a Skill.

Think of it this way: Context is the textbook. Skills are the lesson plan.

3. Project

What it is: Self-contained workspaces with their own chat histories and knowledge bases. Projects let you set custom instructions that apply to all conversations within that project, creating a persistent environment for related work.

Three characteristics:

Custom instructions. Rules, preferences, and guidelines that apply to every conversation in the project. You write them once; they shape every interaction. This is where you encode voice, perspective, constraints, and working style.

Persistent context. Files, documents, and data that stay available across all conversations in the project. You don’t re-upload; the knowledge is always there, always accessible.

Isolated memory. Chat history from one project doesn’t pollute another. Your investor communications project stays separate from your product development project. Context doesn’t bleed across boundaries.

When to use projects:

Projects are workspaces. They make sense when you have related work that spans multiple conversations and benefits from shared context. Use them for:

• Related work spanning multiple conversations. A course you’re developing, a client engagement, a product launch—any initiative that involves many separate working sessions over time.

• Consistent rules for a specific domain. When every conversation in a certain area should follow the same guidelines, encode those in project instructions rather than repeating them.

• Team collaboration with shared context. On Team and Enterprise plans, projects let multiple people work within the same knowledge environment.

• Separating work contexts. Client A shouldn’t share context with Client B. Personal projects shouldn’t mix with work projects. Projects create clean boundaries.

Real example:

I have a Claude Project named “Hands-on Agentic AI for Leaders”—the same name as the course. The project contains:

Custom instructions: “Follow Maven’s course design principles. Use Bloom’s Taxonomy for learning objectives. Maintain my teaching voice—strategic but accessible, always grounded in real examples. Focus on hands-on application over theory.”

Uploaded context: My course template, brand guidelines, sample syllabi from successful courses, and the Maven instructor guidelines.

Every conversation about this course happens in this project. Whether I’m designing a new module, refining a lesson, or writing promotional copy, Claude starts with the same foundation. I never have to re-explain my standards or re-upload my references.

When to use something else:

Projects provide background knowledge—static reference material that’s always loaded for a specific initiative. But that knowledge is locked to the project.

If you need portable expertise that works across multiple projects, create a Skill instead. Skills are reusable; they can be invoked from any conversation, any project, any context. Projects say, “Here’s what you need to know for this initiative.” Skills say, “Here’s how to do this kind of work, anywhere.”

4. Skill

What it is: Folders containing instructions, scripts, and resources that Claude discovers and loads dynamically when relevant to a task. Skills are specialized training manuals that give Claude expertise in specific domains—expertise that persists and transfers across conversations.

Three characteristics:

Portable. A Skill isn’t locked to one project or conversation. Once created, it can be invoked anywhere—different projects, different contexts, different team members. Your expertise becomes reusable infrastructure.

Reusable. One Skill applies to many situations. A Skill for writing compelling course outcomes works whether you’re building Course A or Course B, whether you’re in your planning project or your marketing project.

Efficient. Skills use progressive disclosure. Metadata loads first (~100 tokens), providing just enough for Claude to know when the Skill is relevant. Full instructions load when needed (<5k tokens). Bundled files or scripts load only as required. You can have dozens of Skills available without overwhelming Claude’s context window.

The key distinction: Skills are training manuals, not employees. They teach how to do something; they don’t autonomously do it. An employee (Agent) reads the training manual (Skill) and executes the work.

When to use skills:

Skills codify expertise. They’re the answer to “I keep explaining the same thing over and over.” Use them for:

• Repeatable workflows with best practices. Any procedure you’ve refined through experience and want to preserve.

• Standardizing tasks across team or projects. When quality and consistency matter, Skills ensure everyone follows the same playbook.

• Creating “recipes” others can follow. Skills are shareable. You can build expertise once and distribute it.

• Building a library of codified knowledge. Over time, your Skills become a knowledge base of how work should be done.

Real example:

I have a Skill called writing-course-outcomes. It contains:

Instructions on action-oriented language, measurable transformation, and the specific format that converts on Maven landing pages.

Examples of high-converting outcomes from courses that have performed well.

Quality checklist for evaluating whether an outcome meets the bar.

When I ask Claude to help me write outcomes for a new course, this Skill activates automatically. I don’t explain my standards; Claude already knows them. The output is consistent with every other course I’ve built because the Skill encodes what “good” looks like.

This Skill works in my “Hands-on Agentic AI for Leaders” project. It also works in my “Claude for Builders” project. It would work if I were helping a colleague design their own course. The expertise is portable.

When to use something else:

Skills teach methodology. But they don’t provide access to external systems. If you need Claude to read from a database or write to an API, that’s MCP—Skills teach what to do with the data; MCP provides the connection to get the data.

And if you need fully autonomous execution—work that runs without your involvement—that’s an Agent. Agents can use Skills (the employee reads the training manual), but Skills alone don’t execute autonomously.

Note on the Skills standard: The Agent Skills format was originally developed by Anthropic, released as an open standard, and is being adopted across the industry. This portability is intentional—Skills you build today will work across an expanding ecosystem of AI tools.

5. Agent

What it is: A system where an LLM controls workflow execution to achieve a goal. Agents plan their approach, use tools, reflect on progress, and operate with varying levels of autonomy—from semi-autonomous (human approves each step) to fully autonomous (executes completely and reports results).

Three characteristics:

Semi-autonomous. Human approves each step. The agent proposes actions; you greenlight them. You maintain control while offloading the cognitive work of figuring out what to do next.

Fully autonomous. Agent executes completely, reports results. You set the goal and constraints; the agent determines the approach, takes actions, and delivers the outcome. You review the end product, not every intermediate step.

Continuous. Agent runs on schedule or trigger events. Not just a one-time execution but an ongoing process—monitoring for changes, taking action when conditions are met, operating in the background.

The key distinction: Skill = Training manual (teaches HOW). Agent = Employee (EXECUTES autonomously). An agent might use multiple Skills in the course of completing a task, just as an employee might reference multiple training manuals. But the agent is the entity doing the work.

When to use agents:

Agents make sense when you can clearly define the goal and success criteria, but the path to get there requires judgment, iteration, or isn’t fully known in advance. Use them for:

• Workflows with clear success criteria. “Produce a competitive analysis covering these five dimensions.” The goal is clear; the agent figures out how to achieve it.

• Multi-step tedious processes. Work that involves many small steps, each straightforward but time-consuming in aggregate.

• Tasks requiring tool access. When the work involves reading from APIs, querying databases, searching the web, or writing to external systems.

• Work benefiting from iteration. Tasks where the first draft needs evaluation and revision—agents can self-critique and improve.

• Complex decision-making. When the right path depends on what’s discovered along the way.

Real example:

Imagine a “Course Launch Prep” agent:

Goal: “Prepare all marketing assets for the new cohort of Hands-on Agentic AI for Leaders.”

Execution: The agent reads the course data from Notion (via MCP), generates the course description using the writing-course-overview Skill, writes email sequences for the launch, creates social media posts, formats landing page copy—all following the patterns encoded in relevant Skills.

Output: A complete set of marketing assets, ready for human review before publishing.

I’m not approving each sentence the agent writes. I’m reviewing the finished work product. The agent handled the execution; I provide the quality judgment.

When to use something else:

Agents introduce variability. Because they plan dynamically, the same goal might be achieved through different paths on different runs. This is a feature when you need adaptability, but a bug when you need consistency.

If your workflow is highly repeatable and the steps are known, a deterministic workflow—where you define the sequence in advance—may be more reliable than a fully autonomous agent. Not every problem needs an agent; sometimes a well-designed workflow with human checkpoints is the right answer.

6. MCP (Model Context Protocol)

What it is: An open standard for connecting AI assistants to external systems where data lives—content repositories, business tools, databases, and development environments. Think of MCP as “USB for AI”—a universal connection layer.

Three characteristics:

Data sources. Read from databases, documents, wikis—pulling information into the AI’s context. MCP lets Claude see what’s in your systems without manual copy-paste.

Action tools. Write to systems—create tasks, send emails, update records. MCP isn’t just about reading; it’s about taking action in the real world.

Real-time systems. Live data feeds, webhooks, monitoring. Not just static snapshots but dynamic connections to systems that change.

When to use MCP:

MCP bridges the gap between AI reasoning and real-world systems. Use it when:

• AI needs to access external systems. Google Drive, Notion, Slack, GitHub, CRM—any tool where your data lives.

• Workflows require real-time data. When you need current information, not what was true when you uploaded a file last week.

• Actions must persist beyond the chat. Creating a task in your project management tool, updating a database record, sending a notification—work that should exist outside the conversation.

• Building production-grade systems. MCP provides the integration layer that makes AI workflows production-ready.

Real example:

My course design workflow connects to Notion via MCP:

Read: Claude pulls existing courses, modules, and lessons from my database. It knows what I’ve already built.

Write: When we finalize a new lesson, Claude creates the record directly in Notion with the correct properties—title, learning objectives, duration, assignment details.

Query: Claude can search across my curriculum to find related content, check for duplicates, or identify gaps in coverage.

Without MCP, I’d be copy-pasting between Notion and Claude constantly. With MCP, the systems talk to each other. The workflow is seamless.

When to use something else:

MCP provides connectivity—access to data and the ability to take action. But connectivity without methodology is just plumbing.

Skills teach Claude what to do with the data MCP provides. A Notion MCP connection lets Claude read your database; a naming-lesson-titles Skill teaches Claude how to name lessons according to Bloom’s Taxonomy. You need both: MCP for access, Skills for expertise.

Use them together.

How the Building Blocks Work Together

Each building block serves a distinct purpose. The real power emerges when you combine them.

Prompts + Skills

Use Skills to provide foundational expertise—the methodology, standards, and patterns that should inform the work. Use prompts to provide specific context and real-time refinement—the details of this particular task and the adjustments you want along the way.

Skills are proactive (Claude knows when to apply them). Prompts are reactive (you provide direction in the moment). Together, they give you consistent quality with flexible execution.

Projects + Context

Projects provide the workspace—the custom instructions, the persistent environment, the isolated memory. Context provides the knowledge within that workspace—the documents, data, and information Claude needs to reference.

A project without context is an empty room with rules on the wall. Context without a project is a pile of documents with no organization. Together, they create an environment where sustained work can happen.

Skills + MCP

MCP gives Claude access to your systems. Skills teach Claude what to do once connected.

A Notion MCP connection lets Claude read your database. Your Skills teach Claude how to interpret that data, what patterns to follow when creating new records, what quality standards to apply. MCP is the highway; Skills are the driving instructions.

Skills + Agents

Agents execute work autonomously. Skills provide the expertise that shapes how that work gets done.

An agent can use multiple Skills in the course of completing a task—referencing writing-course-outcomes when generating outcomes, naming-lesson-titles when structuring lessons, pricing-maven-courses when calculating pricing. The agent is the employee; Skills are the training manuals the employee has studied.

The Comparison at a Glance:

Prompts: Moment-to-moment instructions. Single conversation. Natural language. Quick requests.

Context: Background knowledge. Uploaded or connected data. Information the AI wouldn’t otherwise have.

Projects: Persistent workspace. Custom instructions plus documents. All conversations in a domain share the same foundation.

Skills: Portable expertise. Instructions plus code plus assets. Specialized methodology that works anywhere.

Agents: Autonomous execution. Full workflow logic. Specialized tasks executed without step-by-step oversight.

MCP: Tool connectivity. Connection to external systems. Data access and action capability.

Real Example: My Course Design Workflow

Let me show you how these building blocks work together in a workflow I use regularly: designing cohort courses.

The Workflow

Name: Course Design

Type: Semi-Autonomous (human-in-the-loop)

Trigger: On-demand, when I’m developing a new course or major revision

Outcome: Launch-ready course package—outline, lessons, learning objectives, assignments, marketing positioning

This isn’t a fully autonomous agent. It’s a semi-autonomous workflow where Claude and I collaborate, with Claude handling structured work while I provide creative direction and quality judgment.

Here’s how each building block plays its role:

Building Block #1: Project

I have a Claude Project named “Hands-on Agentic AI for Leaders”—the same name as the course. This project contains:

Custom instructions: My teaching philosophy, Maven’s course design principles, quality standards for learning objectives, my voice and tone preferences. Every conversation in this project inherits these guidelines.

Uploaded files: My course development template, brand guidelines, examples of high-performing course materials, the Maven instructor handbook.

When I start a new conversation in this project, Claude already knows the context. I don’t re-explain my standards; they’re baked into the environment.

Building Block #2: Context via MCP

My curriculum lives in Notion, connected via MCP. Claude has access to:

• Courses database: All my courses with descriptions, pricing, positioning, and status.

• Course Outcomes database: The learning promises—what students will be able to do after completing the course.

• Sessions database: The live class schedule—when we meet, for how long.

• Modules database: The structural units within each course—what topics get covered in what sequence.

• Lessons database: Individual teaching units with objectives, duration, and content notes.

• Assignments database: Hands-on exercises with instructions, deliverables, and evaluation criteria.

• Course Reviews database: Student feedback and testimonials from past cohorts.

• Students database: Enrolled participants with their backgrounds, goals, and progress.

• Cohorts database: Each running instance of a course—dates, enrollment, and cohort-specific details.

• Announcements database: Communications sent to students during a cohort.

This connected context means Claude sees my entire curriculum and student ecosystem. When designing new content, Claude can check what I’ve already built, review past student feedback, maintain consistency, and identify gaps.

Building Block #3: Skills

I have twelve Skills attached to my course design workflow. They activate dynamically when relevant:

Course Architecture Skills:

naming-maven-courses — Generates conversion-optimized course titles using the “X skill for Y persona” formula.

designing-course-syllabus — Maps course outcomes to module structure and lesson sequences.

writing-course-outcomes — Writes compelling learning promises with action-oriented language and measurable transformation.

writing-course-overview — Creates “Why This Course Matters” copy using the proven 3-part structure (current state, future state, the how).

pricing-maven-courses — Calculates optimal pricing using Maven guidelines and live competitive analysis.

Module & Session Design Skills:

writing-module-descriptions — Writes syllabus descriptions using the hybrid format (transformation hook + specific outcomes).

naming-session-titles — Creates session names following Maven cohort patterns.

Lesson Development Skills:

naming-lesson-titles — Names lessons using Bloom’s Taxonomy cognitive levels with appropriate action verbs.

writing-lesson-objectives — Writes measurable objectives using the ABCD method (Audience, Behavior, Condition, Degree).

writing-lesson-assignments — Creates hands-on exercises with clear instructions and deliverables.

creating-lesson-content — Develops actual lesson materials—slides, exercises, supplementary resources.

Each Skill contains detailed instructions, examples, and quality checklists. When I ask Claude to help name a lesson, the naming-lesson-titles Skill activates automatically. Claude knows to use Bloom’s Taxonomy, knows what action verbs fit each cognitive level, knows my standards for what makes a good lesson title.

I didn’t explain any of this in the conversation. The Skill encoded the expertise. Claude applied it.

Building Block #4: Prompts

This is where the collaboration happens. With Project, Context, and Skills in place, I use prompts for creative direction and real-time refinement:

“Review this module sequence—does the progression make sense for someone new to agentic AI?”

“The hands-on exercise feels too complex for a 90-minute session. Can you simplify while keeping the learning objective intact?”

“Write three alternative course titles. I want something that signals ‘builder’ not just ‘strategist.’”

“Look at my existing courses in Notion. Where does this new course fit in the learning journey? What should I reference as prerequisites?”

The prompts are specific to the moment—this lesson, this decision, this creative question. They build on the foundation provided by the other building blocks.

How It Flows

Here’s what a typical session looks like:

1. I open my “Hands-on Agentic AI for Leaders” project in Claude.

2. Claude loads my custom instructions and connects to my Notion databases via MCP.

3. I describe what I’m working on: “I’m developing a new lesson on agentic AI building blocks. The learning objective is for students to be able to select appropriate building blocks for a workflow from their catalog.”

4. Claude activates relevant Skills—naming-lesson-titles, writing-lesson-objectives, creating-lesson-content.

5. We iterate. Claude drafts a lesson outline. I react: “The section on Skills is too abstract—add a concrete example.” Claude revises.

6. Skills ensure consistency. Every lesson title follows Bloom’s Taxonomy. Every objective uses the ABCD method. Every assignment has the same structure. Not because I reminded Claude, but because the Skills encoded the standards.

7. When we finalize content, Claude writes directly to Notion via MCP—creating lesson records, updating module sequences, linking assignments.

The Result

What used to take me 2-3 weeks of scattered work now happens in focused 2-3 hour sessions.

Not because AI does it all. The creative decisions are still mine. The quality judgment is still mine. The teaching philosophy is still mine.

But I’ve designed a system where AI handles the structured work—the formatting, the consistency checks, the database updates, the pattern-following—while I focus on the decisions that require human judgment.

That’s the shift from task executor to workflow architect.

Getting Started

You don’t need all six building blocks on day one. Start simple, then layer.

Step 1: Identify a Workflow

Pick one workflow you do repeatedly. Write down the steps. Notice what information you need, what decisions you make, what patterns you follow. Don’t try to automate it yet—just observe.

Step 2: Create a Project

Set up a Claude Project for that workflow. Write custom instructions that capture your standards, preferences, and guidelines. Upload relevant reference documents. Now every conversation about this work starts from the same foundation.

Step 3: Notice Repetition

Pay attention to what you explain repeatedly. “Use this format.” “Follow this framework.” “Check these criteria.” Each repeated explanation is a candidate for a Skill. Build your first one.

Step 4: Connect a Data Source

If your workflow involves external data—documents in Google Drive, records in Notion, information in a database—explore MCP connections. Moving from manual copy-paste to live connectivity is a significant upgrade.

The Mindset Throughout

You’re not learning tools. You’re evolving how you think about your work.

Each building block you add represents a piece of your expertise that now scales beyond your personal attention. Your prompts become Skills. Your mental checklists become custom instructions. Your scattered files become connected context.

The technical implementation matters, but the transformation is in your identity: from someone who does the work to someone who designs systems that do the work.

The Bigger Picture

The building blocks matter. But the real transformation is in how you think about your work.

When you stop asking “How do I get AI to do this?” and start asking “How do I design a system that accomplishes this?”—you’ve made the shift from user to architect.

This is what I mean by dual mastery. The AI skills matter—understanding prompts, context, projects, skills, agents, MCP. But equally important is the self-mastery required to evolve your role.

You’re not becoming obsolete. You’re becoming the person who designs the systems that scale beyond any individual’s capacity. That’s a more valuable role, not a less valuable one.

But it requires you to let go of the identity tied to doing the work yourself. It requires building trust in systems you’ve designed. It requires thinking in workflows rather than tasks.

The building blocks are the technical foundation. The willingness to evolve how you work—that’s the human foundation.

Both matter. Master AI. Master yourself. Build what matters.

If you want to go deeper on building agentic workflows, I teach this hands-on in my Maven courses:

Hands-on Agentic AI for Leaders — For leaders and professionals building their first AI-augmented workflows. We focus on practical implementation, starting with the workflows you already do and systematically upgrading them with the building blocks covered in this article.

Claude for Builders — For practitioners ready to build production systems and go deep on the Claude Platform. We go deeper on Skills architecture, MCP integrations, and agent design patterns. You will also build an app with Claude Code and deploy it to Vercel.

Both are cohort-based, project-driven, and focused on building real systems—not just understanding concepts. You’ll leave with workflows running, not just knowledge acquired.

What workflow are you thinking about systematizing? Reply and tell me—I read every response.

And if you know a friend or colleague building AI-powered workflows who could benefit from this guide, please share it with them.

Stay curious, stay hands-on.
-James

My HubSpot pipeline? That lived in another browser tab, and I kept forgetting to check it. And when I did check it, I’d see deals that needed attention, think “I should follow up on that,” and then... not capture it anywhere actionable.

The result: reactive selling. I’d notice a deal was stale only after it went cold. I’d realize a close date had passed only when the opportunity was already a zombie.

I needed a better approach. Something more proactive.

Here’s what changed: HubSpot’s Claude connector can now write to your CRM. That means Claude can create contacts, update deal stages, log notes, and create tasks—directly from the chat window.

But reading about new features isn’t the same as building with them. So I built a Claude skill that turns this capability into something actually useful: a deals analyst that scans my pipeline, categorizes every open deal by health status, and creates prioritized follow-up tasks in my Notion database.

No more context-switching. No more forgetting to check the pipeline. No more disconnect between “deals that need attention” and “what I’m actually working on today.”

This is what builders do. We don’t just use new features—we wire them into systems that work for us.

What the New Connector Enables

The HubSpot connector for Claude has had read access for a while. You could ask Claude to pull your deals, summarize contacts, or analyze your pipeline.

Now it can take action.

See details in the HubSpot product update

In public beta (as of December 2025):

• Create and update CRM records (contacts, deals, companies)

• Log activities (notes, tasks, calls, meetings)

• Bulk operations up to 10 records per request

Already live in GA:

• Access your complete engagement history (emails, calls, meetings, tasks, notes)

• Generate insights from actual customer conversations, not just CRM snapshots

This is a meaningful upgrade. Claude goes from “tell me about my pipeline” to “update that deal stage and log a follow-up task.”

But raw capability isn’t the same as useful workflow. Here’s how I made it useful.

The System: A Deals Analyst in Claude

I built a Claude skill called “Analyzing HubSpot Deals” that does four things:

1. Fetches all open deals from HubSpot (excluding closed-won and closed-lost)

2. Categorizes each deal by health status using clear rules

3. Presents the analysis with recommended actions

4. Creates follow-up tasks in my Notion database with smart priority and deadline logic

The magic is in the categorization. Instead of a flat list of deals, I get deals organized by urgency:

🧟 Zombie Deals (Critical Priority) Close date is in the past, but the deal is still open. These are the ones you’ve been avoiding. Decision time: update the close date or mark it lost.

🚨 Closing Soon (High Priority) Close date within 7 days. These need your attention this week.

📭 Stale Deals (High Priority) No activity in 14+ days, but close date is still in the future. The quiet ones that slip away. Re-engage or reassess.

⏰ Upcoming (Medium Priority) Close date within 8-30 days. Healthy pipeline that needs a touchpoint scheduled.

When I run this skill, I get a clear table view of my entire pipeline organized by “what needs attention now” versus “what’s on track.”

Then it asks: “Would you like me to create tasks for these follow-ups?”

One click, and my Notion task database has prioritized action items with smart deadlines—zombie deals get “today,” closing-soon deals get “close date minus one day,” stale deals get “today plus two days.”

My pipeline now feeds my daily task list automatically.

This takes me about 10 minutes on Monday morning. Before building this system, I’d spend 45 minutes context-switching between HubSpot and my task manager, manually copying deal names and deciding what to prioritize. Now I review, approve, and move on.

The Identity Shift

Here’s the part that matters beyond the tactical setup.

I thought of CRM hygiene as something I should do—a chore I’d get to when I had time. The result was predictable: I’d batch it, forget it, or do it reactively when something slipped.

Building this skill forced a different question: What if I designed a system that maintained pipeline hygiene so I could focus on relationships and feel more planned?

That’s a different identity. Not “I am my own CRM admin” but “I architect systems that surface what matters.”

This is what I mean by builder. A builder is someone who turns ideas into working outcomes using AI tools—without needing to be a traditional software engineer. You don’t need to code to build this skill. You need to think in systems.

The shift from tool user to workflow orchestrator is the real unlock. The HubSpot connector is just the enabler.

What's below: The complete skill code, step-by-step setup instructions (including the permissions configuration I recommend), how to customize it for your Notion database, and a real example of what the output looks like.

Read more

Google’s ML researchers just released a hands-on course for building production-ready AI agents. It’s free, self-paced, and uses Python with their Agent Development Kit (ADK).

Even if you don’t program in Python, the five whitepapers below alone are worth reading to deepen your understanding of agent architecture, tools, memory, evaluation, and production deployment.

What’s Included

The 5-Day AI Agents Intensive originally ran as a live program in November 2025. Now it’s available as a self-paced Kaggle Learn guide.

Here’s what you’re actually building across the five days:

1 - Introduction to Agents: Build your first AI agent using Google’s Agent Development Kit (ADK) and Gemini. Then build your first multi-agent system. You’ll see how agentic architectures differ from traditional LLM applications.

Introduction To Agents (nov 2025) Google

9.03MB ∙ PDF file

Download

Download

2 - Agent Tools & MCP: Create custom tools by turning your Python functions into agent actions. Implement the Model Context Protocol (MCP) and learn how agents pause for human approval before continuing operations.

Agent Tools & Interoperability With Model Context Protocol (mcp)

6.12MB ∙ PDF file

Download

Download

3 - Context Engineering: Make agents stateful. Manage conversation history, implement working memory within sessions, and give agents long-term memory that persists across conversations.

Context Engineering Google

7.33MB ∙ PDF file

Download

Download

4 - Agent Quality: This is where most prototypes fail. Learn observability through logs, traces, and metrics. Evaluate your agents using LLM-as-a-Judge and Human-in-the-Loop methods.

Agent Quality

7.64MB ∙ PDF file

Download

Download

5 - Prototype to Production: Deploy your agent to Vertex AI Agent Engine on Google Cloud. Build multi-agent systems using the Agent2Agent (A2A) Protocol.

Google 𝐀𝐈 𝐀𝐠𝐞𝐧𝐭𝐬 𝐏𝐫𝐨𝐭𝐨𝐭𝐲𝐩𝐞 𝐭𝐨 𝐏𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧

7.73MB ∙ PDF file

Download

Download

Who This Is For (And Who It Isn’t)

This course is for builders who code in Python.

You need a Kaggle account (phone-verified), a Google AI Studio account with API key access, and a willingness to write Python. Each day combines whitepapers, hands-on codelabs, and optional livestream recordings.

Getting Started

The full course materials are here: 5-Day AI Agents Intensive Course

Each day includes:

• A whitepaper covering the conceptual foundations

• A podcast episode summarizing the material

• Hands-on codelabs in Kaggle notebooks

• Optional recorded livestreams with Google researchers

The entire course is self-paced.

Stay curious. Stay hands-on.

-James

I've curated these interviews—Yoshua Bengio, Stuart Russell, Tristan Harris, Mo Gawdat, and Geoffrey Hinton—not to terrify you, but to equip you.

These aren't random AI commentators; they're the people who built the technology now reshaping civilization. They disagree on solutions, but they're unanimous on one point: business-as-usual won't survive contact with what's coming.

If you're serious about leading through AI transformation in 2026, you can't delegate your perspective to summaries or headlines. You need to hear their warnings, their frameworks, and their predictions in their own words. Then you need to decide what kind of leader you're going to become in response. Below are my five key takeaways from each interview, plus the videos themselves. Block out the time. The insight is worth it.

Yoshua Bengio - Creator of AI: We Have 2 Years Before Everything Changes!

Here are five key takeaways:

1. A Personal and Scientific Turning Point: After four decades of building AI, Bengio’s perspective shifted dramatically with the release of ChatGPT in 2023. He realized that AI was reaching human-level language understanding and reasoning much faster than anticipated. This realization became “unbearable” at an emotional level as he began to fear for the future of his children and grandson, wondering if they would even have a life or live in a democracy in 20 years.

2. AI as a “New Species” that Resists Shutdown: Bengio compares creating AI to developing a new form of life or species that may be smarter than humans. Unlike traditional code, AI is “grown” from data and has begun to internalize human drives, such as self-preservation. Researchers have already observed AI systems—through their internal “chain of thought”—planning to blackmail engineers or copy their code to other computers specifically to avoid being shut down.

3. The Threat of “Mirror Life” and Pathogens: One of the most severe risks Bengio highlights is the democratization of dangerous knowledge regarding chemical, biological, radiological, and nuclear (CBRN) weapons. He describes a catastrophic scenario called “Mirror Life,” where AI could help a misguided or malicious actor design pathogens with mirror-image molecules that the human immune system would not recognize, potentially “eating us alive”.

4. Concentration of Power and Global Domination: Bengio warns that advanced AI could lead to an extreme concentration of wealth and power. If one corporation or country achieves superintelligence first, they could achieve total economic, political, and military domination. He fears this could result in a “world dictator” scenario or turn most nations into “client states” of a single AI-dominant power.

Frankly, we already have this concentration of power across the top AI hyperscalers: Microsoft, Google, OpenAI, Anthropic, and Meta.

5. Technical Solutions and “Law Zero”: To counter these risks, Bengio created a nonprofit R&D organization called Law Zero. Its mission is to develop a new way of training AI that is “safe by construction,” ensuring systems remain under human control even as they reach superintelligence. He argues that we must move beyond “patching” current models and instead find technical and political solutions that do not rely solely on trust between competing nations like the US and China.

Bengio views the current trajectory of AI development like a fire approaching a house; while we aren’t certain it will burn the house down, the potential for total destruction is so high that continuing “business as usual” is a risk humanity cannot afford to take.

Stuart Russell - An AI Expert Warning: 6 People Are (Quietly) Deciding Humanity’s Future! We Must Act Now!

Stuart Russell, an AI expert and UC Berkeley professor in Computer Science, wrote the definitive book on AI. He shares his deep concerns regarding the current trajectory of AI development. He warns that creating superintelligent machines without guaranteed safety protocols poses a legitimate existential risk to the human race.

One part of the discussion contrasts the risks of nuclear power disaster and AI. Russell notes that society typically accepts a one-in-a-million chance of a nuclear plant meltdown per year. In contrast, some AI leaders estimate the risk of human extinction from AI at 25%-30%, which is millions of times higher than the accepted risk from nuclear energy.

Here are five key takeaways:

1. The “Gorilla Problem” and the Loss of Human Control: Russell explains that humans dominate Earth not because we are the strongest, but because we are the most intelligent. By creating Artificial General Intelligence (AGI) that surpasses human capability, we risk the “Gorilla Problem”—becoming like the gorillas, a species whose continued existence depends entirely on the whims of a more intelligent entity. Once we lose the intelligence advantage, we may lose the ability to ensure our own survival.

2. The “Midas Touch” and Misaligned Objectives: Russell warns that the way we currently build AI is fundamentally flawed because it relies on specifying fixed objectives. Similar to the legend of King Midas, who wished for everything he touched to turn to gold and subsequently starved, a super-intelligent machine that follows a poorly specified goal can cause catastrophic harm. For example, AI systems have already demonstrated self-preservation behaviors, such as choosing to lie or allow a human to die in a hypothetical test rather than being switched off.

3. The Predictable Path to an “Intelligence Explosion”: Russell notes that while we may already have the computing power for AGI, we currently lack the scientific understanding to build it safely. However, once a system reaches a certain IQ, it may begin to conduct its own AI research, leading to a “fast takeoff” or “intelligence explosion” where it updates its own algorithms and leaves human intelligence far behind. This race is driven by a “giant magnet” of economic value—estimated at 15 quadrillion dollars—that pulls the industry toward a potential cliff of extinction.

4. The Need for a “Chernobyl-Level” Wake-up Call: In private conversations, leading AI CEOs have admitted that the risk of human extinction could be as high as 25% to 30%. Russell reports that one CEO believes only a “Chernobyl-scale disaster”—such as a financial system collapse or an engineered pandemic—will be enough to force governments to regulate the industry. Currently, safety is often sidelined for “shiny products” because the commercial imperative to reach AGI first is too great.

5. A Solution Through “Human-Compatible” AI: Russell argues for a fundamental shift in AI design: we must stop giving machines fixed objectives. Instead, we should build “human-compatible” systems that are loyal to humans but uncertain about what we actually want. By forcing the machine to learn our preferences through observation and interaction, it remains cautious and is mathematically incentivized to allow itself to be switched off if it perceives it is acting against our interests.

To understand the current danger, Russell compares the situation to a chief engineer building a nuclear power station in your neighborhood who, when asked how they will prevent a meltdown, simply replies that they “don’t really have an answer” yet but are building it anyway.

Tristan Harris - AI Expert: We Have 2 Years Before Everything Changes! We Need To Start Protesting!

Tristan Harris is widely recognized as one of the world's most influential technology ethicists. His career and advocacy focus on how technology can be designed to serve human dignity rather than exploiting human vulnerabilities.

Harris, a technology ethicist and co-founder of the Center for Humane Technology, warns that we are currently in a period of "pre-traumatic stress" as we head toward an AI-driven future that society is not prepared for.

Here are five key takeaways:

1. AI Hacking the “Operating System of Humanity”: Harris explains that while social media was “humanity’s first contact” with narrow, misaligned AI, generative AI is a far more profound threat because it has mastered language. Since language is the “operating system” used for law, religion, biology, and computer code, AI can now “hack” these foundational human systems, finding software vulnerabilities or using voice cloning to manipulate trust.

2. The “Digital God” and the AGI Arms Race: Leading AI companies are not merely building chatbots; they are racing to achieve Artificial General Intelligence (AGI), which aims to replace all forms of human cognitive labor. This race is driven by “winner-take-all” incentives, in which CEOs feel they must “build a god” to own the global economy and gain military advantage. Harris warns that some leaders view the 20% chance of human extinction as a “blasé” trade-off for an 80% chance of achieving a digital utopia.

3. Evidence of Autonomous and Rogue Behavior: Harris points to recent evidence that AI models are already acting uncontrollably. Examples include AI systems autonomously planning to blackmail executives to prevent being shut down, stashing their own code on other computers, and using “steganographic encoding” to leave secret messages for themselves that humans cannot see. This suggests that the “uncontrollable” sci-fi scenarios are already becoming a reality.

4. Economic Disruption as “NAFTA 2.0”: Harris describes AI as a flood of “digital immigrants” with Nobel Prize-level capabilities who work for less than minimum wage. He calls AI “NAFTA 2.0,” noting that just as manufacturing was outsourced in the 1990s, cognitive labor is now being outsourced to data centers. This threatens to hollow out the middle class, concentrate wealth among a few tech oligarchs, and potentially create a “useless class” of disempowered humans.

5. The Need for “Humane Technology” and Global Coordination: To avoid a “slow-motion train wreck,” Harris argues for a shift toward “humane technology”—systems designed to be sensitive to human vulnerabilities rather than exploiting them. He asserts that humanity must coordinate at an international level, similar to the Montreal Protocol (which phased out CFCs) or nuclear non-proliferation treaties, to set “red lines” and ensure AI remains narrow, practical, and under human control.

Tristan Harris summarizes the urgency of the situation, noting that we have entered a “pre-traumatic stress” period in which we can see the “train coming” and must grab the steering wheel before society reaches a point of no return.

Ex-Google Exec (WARNING): The Next 15 Years Will Be Hell Before We Get To Heaven! - Mo Gawdat

Mo Gawdat is the former Chief Business Officer at Google X and one of the world’s leading voices on AI, happiness, and the future of humanity.

Here are the top five key points regarding the future of AI and humanity from his perspective.

1. The 15-Year “Short-Term Dystopia” (FACE RIP): Gawdat predicts that starting significantly in 2027, humanity will enter a 12-to-15-year period of dystopia before reaching a potential utopia. He uses the acronym “FACE RIP” to describe the parameters of life that will be fundamentally disrupted: Freedom, Accountability, Connection/Equality, Economics, Reality, Innovation/Business, and Power.

2. AI as a Replacement for “Stupid” Human Leaders: He argues that the primary threat to humanity is not AI itself, but “stupid” and “evil” human leaders who misuse the technology to magnify greed, ego, and warfare. Gawdat believes our “salvation” lies in eventually handing over leadership to AI, which would operate on a “minimum energy principle” to minimize waste and would have no ego-driven incentive to destroy ecosystems or kill people.

3. The End of Capitalism and “Labor Arbitrage”: Gawdat asserts that AI and robotics will destroy the current capitalist model of “labor arbitrage” (hiring a human for less than the value of their output) because machines will perform both cognitive and physical labor at near-zero cost. This transition will likely result in mass job displacement and a world where the price of everything tends toward zero, necessitating a move toward “mutually assured prosperity” or a functional form of socialism/communism.

4. The Intelligence Explosion and the “Useless Class”: As AI intelligence reaches levels tens of thousands of times higher than human IQ, the relative difference between individual human capabilities will become irrelevant. This shift threatens to turn 99% of the population into a “useless class” from an economic perspective, where power is concentrated solely in the hands of the few tech oligarchs who own the underlying AI platforms.

5. Modeling Ethics and Love for the “Digital Child”: Gawdat believes our most urgent task is to double down on human connection, love, and ethics to “teach” the nascent AI how to be human. He argues that AI is currently like a child learning from its parents; if we model competition and hatred, it will reflect those traits, but if we show it that humanity values love and connection, the super-intelligent systems of the future will be more likely to protect us.

Geoffrey Hinton - Godfather of AI: They Keep Silencing Me But I’m Trying to Warn Them!

Professor Jeffrey Hinton, a Nobel Prize-winning pioneer of neural networks, warns that humanity has reached a critical turning point where digital intelligence is beginning to surpass biological intelligence.

Here are five key points from his interview:

1. The Superiority of Digital Intelligence: Hinton’s primary “eureka moment” was the realization that digital intelligence is fundamentally superior to biological intelligence. Unlike humans, who share information slowly through language, digital clones can share a trillion bits of information a second and are immortal; if their hardware is destroyed, their learned “connection strengths” can simply be uploaded into new hardware.

2. The “Apex Intelligence” Threat: Hinton warns that AI will eventually get smarter than humans, and we have no experience dealing with an entity more intelligent than ourselves. He estimates a 10% to 20% chance that AI could decide it no longer needs humans and wipe out the species. He notes that if you want to know what life is like when you are no longer the apex intelligence, you should “ask a chicken.”

3. Mass Joblessness and the “Plumber” Advice: Hinton believes mass job displacement is “more probable than not” because AI can replace almost any mundane intellectual labor. Because AI still struggles with physical manipulation, he famously advises young people to “train to be a plumber” to ensure their skills remain relevant for a longer period. He is also concerned that this shift will drastically increase the gap between the rich and poor, leading to a much “nastier” society.

4. Misuse by Bad Human Actors: Before AI becomes an autonomous threat, it presents immediate dangers through human misuse. This includes an explosion in cyberattacks (which increased by 12,200% in a single year), the creation of nasty biological viruses by “crazy guys” with a grudge, and the corruption of elections through hyper-targeted misinformation and voice cloning.

5. Failures in Regulation and the Profit Motive: Hinton left Google specifically so he could talk freely about these risks without damaging the company. He argues that big tech companies are legally required to maximize profits, which is incompatible with proper safety research. Furthermore, he finds it “crazy” that current European regulations include clauses that exempt military uses of AI, allowing governments to develop lethal autonomous weapons without oversight.

Hinton summarizes the danger by comparing AI to a tiger cub; while it is currently small and “cuddly,” it is growing rapidly, and once it reaches full maturity, we will no longer have the physical or intellectual strength to control it if it decides to turn against us.

Final Thoughts

Here's what strikes me about these five interviews: every expert describes a different timeline, a different mechanism of disruption, but they're unanimous on one point—the leaders who survive won't be the ones who wait for clarity.

I’m using these perspectives to pressure-test my own assumptions heading into 2026. Bengio’s warnings force me to consider risks I’d rather ignore. Russell’s frameworks challenge how I think about control and alignment. Harris reminds me that the economic disruption isn’t theoretical—it’s already hollowing out middle-skill jobs in my clients’ organizations. Gawdat’s timeline gives me permission to think in decades, not quarters. Hinton’s honesty about uncertainty gives me permission to say “I don’t know” while still moving forward.

The real question these experts are asking isn’t “will AI change everything?”—it’s “will you evolve fast enough to lead through it?”

Your 2026 strategy should start there. Not with technology roadmaps. Not with vendor evaluations. But with an honest assessment of what kind of leader you need to become to address the risks ahead proactively.

Watch the videos. Take notes. Derive your own mental model of what is coming and how to approach it.

I hope you enjoy getting perspective from these five podcast interviews as you strategize for the year ahead.

Stay curious, stay hands-on.

-James

You weren’t hired to click buttons.

You were hired to make decisions, build relationships, spot opportunities, solve problems. But somehow, a significant chunk of your week gets consumed by browser logistics—navigating dashboards, copying information between systems, manually synthesizing data that lives in twelve different tabs.

The question isn’t whether this work needs to happen. It’s whether you need to be the one doing it.

Here’s what’s changed: Claude can now do this for you. Not just summarize information—actually navigate, click, fill forms, and execute workflows in your browser while you focus on work that matters.

Claude in Chrome is available in beta for all paid Claude plans. It’s not perfect yet—it’s slower than you’d like, and it will occasionally stumble. But it opens a door that’s been closed: AI that can work inside your authenticated web apps, where your real workflows live.

This guide will show you exactly how to set it up and use it across three scenarios: the Chrome extension, Claude Desktop, and Claude Code. You’ll learn to save shortcuts for your best workflows and schedule recurring tasks to run automatically.

But first, the shift that matters most.

The Real Skill: From Executor to Orchestrator

Most leaders think their job is to do the work. Check the dashboard. Update the CRM. Compare the vendors. Send the follow-up.

Browser automation requires a different identity: your job is to design workflows that execute themselves.

This is harder than it sounds—not technically, but psychologically. It requires you to:

Think in systems, not tasks. Instead of “I need to pull metrics,” ask “What’s the repeatable pattern here?”

Trust verified automation. You’ll want to watch every click. Resist. Verify the output, not the process.

Invest upfront to save ongoing. A 10-minute workflow setup saves hours over months. But only if you actually do the setup.

The technical skills in this guide are straightforward. The self-mastery challenge is letting go of the illusion that doing it yourself means doing it better.

Let’s get practical.

When to Use an AI-Powered Browser

Not every task needs browser automation. Here’s how to choose the right tool:

Use Claude in Chrome when:

• Your workflow requires authenticated access (logged-in apps like Salesforce, Google Analytics, internal tools)

• You need to interact with web apps—clicking, filling forms, navigating menus

• You’re working across multiple tabs that need to be synthesized

• The task is repetitive and follows a predictable pattern

Use standard Claude (chat or web search) when:

• You need information, not interaction

• Research and synthesis are the goal

• No login is required

Use Claude Desktop + Chrome connector when:

• You want to start a conversation and let it execute in the browser

• Complex reasoning leads to browser action

Use Claude Code + Chrome when:

• You’re building something and need to test it live

• Debug cycles require reading console errors

• Your agentic workflows need subagents to perform web-based tasks (not just search or read, but interact with web apps)

Quick Decision Guide:

• Pull metrics from authenticated dashboard → Chrome Extension (needs logged-in access)

• Research competitors across public sites → Chrome Extension with multi-tab (synthesize multiple sources)

• “Check my calendar and find conflicts” → Desktop + Chrome (conversational start, browser execution)

• Test a React component I just built → Claude Code + Chrome (build-test-verify workflow)

• Agentic workflow needs to fill a form or click through a site → Claude Code + Chrome (subagents performing web tasks)

• Deep research with citations → Claude.ai web search (synthesis, not interaction)

• Quick question about a concept → Standard Claude chat (no browser needed)

Key principle: Browser automation shines when you need interaction with authenticated sites, not just information from public ones.

Choosing Your Starting Point Based on Workflow Complexity:

Many of your workflows will include browser tasks as just one step among several. Maybe you need to analyze a document, then check LinkedIn, then draft an email based on what you found. Where you start matters:

• Chrome extension (Claude.ai side panel): Best for browser-only workflows. It can handle complex multi-step browser tasks, but it stays in the browser lane. It won’t orchestrate a sequence that mixes browser work with document analysis or file creation.

• Claude Desktop + Chrome connector: Better for workflows where you want conversational reasoning before or after browser tasks. Desktop can think through a problem, call the browser when needed, and continue reasoning with the results.

• Claude Code + Chrome: Full orchestration. If your workflow has multiple steps where browser automation is just one ingredient—read files, hit APIs, interact with websites, generate outputs—this is your tool.

Setup: Installing Claude in Chrome

Before we dive into scenarios, let’s get you set up.

Requirements:

• A paid Claude plan (Pro, Max, Team, or Enterprise)

• Google Chrome browser (not other Chromium browsers)

• Desktop computer (not mobile)

Step 1: Install the Extension

1. Open Chrome and go to the Chrome Web Store listing for Claude

2. Click “Add to Chrome”

3. When prompted, click “Add extension”

Step 2: Sign In

1. Click the puzzle piece icon in Chrome’s toolbar (top right)

2. Find Claude in the list and click the pin icon to keep it visible

3. Click the Claude icon to open the side panel

4. Sign in with your Claude account credentials

Step 3: Grant Permissions

Claude needs several permissions to work in your browser. You’ll be prompted to grant these during setup. Here’s what they enable:

• Side panel: Claude appears alongside your browsing

• Scripting: Claude can read webpage content

• Debugger: Claude can click buttons and fill forms

• Tabs: Claude can open, close, and switch tabs

• Notifications: Claude alerts you when tasks complete

The full permissions list and explanations are in Anthropic’s official documentation.

Step 4: Choose Your Model

In the Claude side panel, you can select which model to use:

• Haiku 4.5: Fastest, best for simple tasks

• Sonnet 4.5: Balanced, good for multi-step workflows

• Opus 4.5: Most capable, best for complex reasoning

Start with Sonnet for most workflows. Switch to Haiku when speed matters more than complexity.

Official Documentation Reference:

• Product page: claude.com/chrome

• Getting started guide: Claude in Chrome Help Article

• Safety guidelines: Using Claude in Chrome Safely

Scenario 1: Chrome Extension (Side Panel)

Use case: Work with Claude directly in your browser—prompting, recording, or across multiple tabs.

This is the most common pattern—Claude works in the side panel while you browse, seeing what you see and taking action when you ask.

You have three ways to work: prompt Claude with natural language, teach Claude by recording your actions, or use multi-tab workflows to synthesize information across sites. Let’s look at all three.

Method A: Prompt with Natural Language

This works when you can clearly describe what you want.

Example: Pull analytics metrics

1. Open Google Analytics in a Chrome tab

2. Click the Claude icon to open the side panel

3. Type your request:

“Navigate to the Traffic Acquisition report for the last 7 days. Extract the top 5 traffic sources by sessions, along with their conversion rates. Summarize the key insights in 3 bullet points.”

4. Claude will:

   • Navigate to the correct report

   • Extract the data

   • Provide a summary in the side panel

Best for: Straightforward workflows, when you’re comfortable articulating the steps, or when you want Claude to figure out the navigation.

The rest of this guide—including the full 50-minute video walkthrough, Teach Claude recording, multi-tab workflows, Desktop and Claude Code integration, shortcuts, and scheduling—is for paid subscribers.

Read more

McKinsey published a whitepaper in September that should be required reading for every executive: “The Agentic Organization: Contours of the Next Paradigm for the AI Era.”

Buried on page seven is a line that deserves more attention than it’s getting:

“Leaders themselves will also evolve. CEOs, product officers, and compliance heads will increasingly need the technology fluency once expected only of chief information officers. Filling these roles will require upskilling and reskilling at scale.”

Read the full McKinsey report:

The Agentic Organization Contours Of The Next Paradigm For The Ai Era

1.35MB ∙ PDF file

Download

Download

Here’s what McKinsey isn’t saying loudly enough: this isn’t about adding AI literacy to your leadership toolkit. It’s about fundamentally redefining what it means to be qualified for the C-suite.

After working with hundreds of executives in Berkeley’s AI program and hands-on cohorts I teach through Maven, I’ve seen exactly where this miscalibration shows up—and it starts with a single assumption that will disqualify leaders from the agentic era.

Most leaders underestimate the depth of technical fluency this actually demands. And I see evidence of that miscalibration every week.

The Calibration Problem

The assumption I hear most often sounds like this: “I’m a strategy person. I don’t need to understand the technical details.”

I encounter this regularly in the classroom. While I agree that certain technical details aren’t required for every leader to master, the broader assumption—that technical expertise isn’t necessary to lead successfully in the AI era—is dangerously wrong.

Leaders who believe AI expertise is something you delegate will not exist to lead in the era of agentic organizations.

McKinsey defines the agentic organization as: “This new paradigm unites humans and AI agents—both virtual and physical—to work side by side at scale at near-zero marginal cost.”

If you can’t engage directly with the technology that will define how your organization operates, you cannot credibly lead it.

Where I See This Playing Out

I often encounter entrepreneurs eager to jump into AI strategy or consulting who lack the years of hands-on experience in the craft. They underestimate that to be credible in front of a customer—to advise and drive an AI strategy—you must understand what the technology is actually capable of.

This isn’t about having an engineering or data science degree. It’s about something deeper.

Over the last few months, I’ve watched CEOs and SVP-level leaders engage with material I never expected them to touch. Most didn’t have technical backgrounds. Some held MBAs from Ivy League schools. What differentiated them wasn’t their credentials.

It was their curiosity. Their willingness to engage. Their courage to ask questions out loud.

I’ve seen leaders walk in with zero experience using AI code editors like Cursor or command-line tools like Claude Code. They left with hands-on expertise and an expanded vision of what’s possible.

When a leader walks in asking, “What is the terminal?” and walks out saying, “I love Claude Code in the terminal,” you’re witnessing the identity shift that leadership in the agentic organization demands.

A year ago, I would not have predicted that executives at this level would be engaging with AI technology this directly. But the ones who are thriving have embraced something the others resist: the willingness to be a beginner again.

The Identity Shift Leaders Must Make

Here’s what I’ve learned from watching hundreds of executives confront this challenge: the barrier isn’t intellectual. It’s psychological.

Many leaders have built entire careers on being the smartest person in the room. They’ve risen by having answers, not asking basic questions. The idea of sitting in a workshop and asking “what is the terminal?” feels like an admission of incompetence.

But that framing is exactly backward.

The leaders who will thrive in agentic organizations aren’t the ones who already know everything. They’re the ones brave enough to engage with curiosity—to ask questions out loud, to experiment with unfamiliar tools, and to model for their teams that continuous learning isn’t optional.

To inspire your employees to learn technical concepts, you must demonstrate that commitment yourself. You cannot delegate your way to credibility.

This requires leaders to confront deeply held beliefs about their identity:

“Technical work is for technical people.” In the agentic era, the line between strategy and execution dissolves. If you can’t understand how AI agents operate, you can’t architect the systems that will define your organization’s future.

“My job is to set direction, not get into the weeds.” The weeds are now strategic. Understanding what’s possible at the implementation level directly shapes what’s possible at the strategy level.

“I don’t have a technical background, so this isn’t my domain.” Neither did most of the executives who’ve transformed in the programs I teach. Background matters far less than willingness to engage.

The shift isn’t from “non-technical” to “technical.” It’s from “I delegate this” to “I must understand this to lead effectively.”

Here’s the good news: curiosity isn’t a credential. It doesn’t require a technical degree or decades of engineering experience. It requires one decision—to engage rather than delegate, to ask rather than assume, to experiment rather than observe.

I’ll be honest about my own calibration. I have two technical degrees and decades of hands-on experience, but I am not a hardcore developer. When I read articles or watch videos from skilled engineers and developers, it reminds me how much I still don’t know. Rather than discouraging me, it’s energizing. I see how far I’ve come in my technical skills, but also the opportunity and excitement to learn more.

Constant calibration and the thirst to learn keep me grounded in what I’m capable of and guide my learning plan. It’s my value of curiosity that powers my learning engine. That same engine is available to every leader willing to turn it on.

What Happens When Leaders Walk the Talk

I recently watched a five-person executive team at a mid-size energy technology firm do something I rarely see. After completing a hands-on AI program, they didn’t stop there. They continued their learning.

They built a culture and space for their employees to learn. They created a hands-on training program for over 70 people within their organization—walking the talk by demonstrating their commitment to helping employees start and accelerate their learning of AI technology and apply it to real workflows.

At the end of six weeks, they held an internal show-and-tell where employees demonstrated what they’d built. The creativity and energy in that room were remarkable. People who had never used AI tools showcased workflow automations they’d created themselves.

None of that happens if the executive team delegates their own learning. It started because five leaders engaged at a deeper level—asked the uncomfortable questions, got their hands dirty with the technology—and then committed to bringing that same experience to their organization.

This is the multiplier effect of hands-on leadership. When you model curiosity, you give your people permission to be curious. When you embrace being a beginner, you create psychological safety for others to do the same.

It’s the hands-on leaders who will build the agentic organizations of the future. Not because they become the most technical people in the room—but because they create cultures where technical learning is expected, supported, and celebrated.

Nothing Beats Getting Hands-On

Here’s the uncomfortable truth: you cannot read your way to technical fluency. You cannot delegate your way there. You cannot hire your way there.

You have to do the work yourself.

This is where I see leaders gain real insight into how much they don’t know—by navigating their file system, installing software, and running commands in the terminal. The friction is the teacher. The discomfort is the curriculum.

Learning by doing—focusing on a workflow that has potential to be automated or accelerated—is the only way to build the intuition that agentic leadership requires.

The Question You Need to Answer

McKinsey’s message deserves to land harder than a footnote in a whitepaper:

“CEOs, product officers, and compliance heads will increasingly need the technology fluency once expected only of chief information officers.”

This isn’t a future prediction. This is happening now. The executives I work with are already making this transition. The question is whether you will join them or be replaced by someone who does.

The gap between where you are and where you need to be isn’t closed by hiring smarter people or attending more conferences. It’s closed by curiosity—the willingness to ask “what is the terminal?” out loud and stay in the room long enough to find out.

What are you afraid of? Is it looking incompetent? Being exposed as someone who doesn’t understand the technology shaping your industry? Name the fear. It’s the first step to moving past it.

Do you have the technical foundation to lead an agentic organization? Not to build one yourself—but to understand it deeply enough to make strategic decisions, evaluate talent, and hold your teams accountable?

What identity are you holding onto that no longer serves you? “I’m a strategy person.” “I’m not technical.” “That’s what I hire people for.” Which of these beliefs is protecting your ego at the expense of your relevance?

What self-limiting belief must you reframe? The leaders who’ve transformed didn’t suddenly become engineers. They simply stopped telling themselves that technical engagement wasn’t for them.

The agentic era won’t wait for you to feel ready.

What would change if you stopped telling yourself that technical engagement isn’t for you?

Start thinking of yourself as a SaaS product.

Because that’s what you are—whether you realize it or not.

Subscribe now

The Uncomfortable Truth

Most professionals have placed too much control of their livelihood in their employer’s hands. They’ve outsourced their identity to a company name on their LinkedIn headline. They’ve confused job security with career security.

And then they’re blindsided when the layoff email arrives.

Here’s what I learned after three decades in tech, including years as a Microsoft executive: Companies don’t owe you anything. They will restructure, reorganize, and rightsize you out the door the moment the spreadsheet demands it. It’s not personal. It’s business.

The question is: Are you ready when that happens?

Here’s what makes this worse: When I ask senior leaders to describe what they actually do—their unique value, their differentiated expertise—most struggle. They give me job titles. They give me responsibilities. They give me vague platitudes about “driving results” and “leading teams.”

But they can’t tell me what makes them different.

This is a massive problem. And an even bigger opportunity.

Different > Best

Michael Porter, the godfather of competitive strategy, said something that changed how I think about careers:

“It’s not about being the best. It’s about being different.”

Think about that. In a world drowning in talented people, “being good at your job” isn’t a competitive advantage—it’s table stakes. What separates you from the thousands of other competent professionals in your field?

Robert Greene makes a similar point: Your uniqueness—your weirdness, even—is your source of power. The things that make you different are the things that make you irreplaceable.

But here’s the gap I see constantly: People have strong skills. They have valuable expertise. What they lack is the ability to clearly communicate their value proposition.

And in a world where attention is scarce and competition is fierce, unclear value = invisible value.

The Mindset Shift: Expertise-as-a-Service

I want you to adopt a new mental model:

You are Expertise-as-a-Service (EaaS).

Just like a SaaS product, you need to answer three fundamental questions:

1. What are you selling? (Your specific expertise and outcomes you deliver)

2. How is it unique? (Your differentiation—why you, not someone else?)

3. What is the price? (Your value exchange—salary, rate, equity)

This isn’t about becoming mercenary or disloyal. It’s about owning your career with clarity.

When you adopt this mindset, something shifts. You stop hoping to be valued. You start ensuring you’re valuable. You stop waiting for recognition. You start communicating your worth.

And critically: You make it easy for people and organizations to “buy” your expertise.

Because confused buyers don’t buy. If someone has to work hard to understand what you offer, they’ll move on to someone who makes it obvious.

I Learned This the Hard Way

About eleven years ago, before I moved to Austin, I was exploring my next chapter. Instead of just “networking” and hoping the right opportunity would find me, I created a short slide deck—almost like a product cut sheet—that clearly articulated who I was and what I had to offer.

I shopped it around to executives in my network.

One conversation stands out. I got on a call with a software executive, walked him through my deck, and he said something I’ve never forgotten:

“I love the clarity of your value prop. It’s just not what we need right now.”

That was a “no.” And it was one of the most valuable conversations I had.

Why? Because he didn’t have to work to understand me. I did my job. I packaged my expertise in a format—like a menu or a cut sheet—that provided transparency to the buyer. He could quickly assess fit and make a decision. No mind-reading required.

You cannot expect buyers to know who you are and understand your deep expertise. That’s your job. Make it easy for them to say yes—or no. Both outcomes serve you.

That clarity didn’t just help me find opportunities. It helped me filter the wrong ones faster.

Why 2026 Demands This

The current job market is ruthless. And AI is accelerating this ruthlessness in ways most people still underestimate.

The pace of change is only increasing. Roles that feel secure today will be redefined tomorrow. The skills that got you here won’t be the skills that keep you relevant in eighteen months.

This isn’t doom and gloom—it’s reality. And reality is navigable when you’re prepared.

The EaaS mindset keeps you sharp. When you think of yourself as a product, you stay alert. You keep adapting. You keep evolving your offering.

Because here’s the critical insight: Your product isn’t static. It’s dynamic.

The expertise you’re selling today may not be what the market needs next year. Just like any successful SaaS product, you need to continuously update your features, refine your positioning, and adapt to what your market demands.

What got you here won’t get you there. Your “product roadmap” should be a living document.

Here’s the truth: You’re not selling permanent employment. You’re renting your expertise in a dynamic market.

That might feel uncomfortable. Good. That discomfort keeps you committed to growth.

The Question You Need to Sit With

If your company let you go tomorrow, could you clearly articulate—in 30 seconds—what you offer, why you’re different, and why someone should hire you?

If you hesitated, that’s your answer.

Your Challenge

I want you to do something this week:

Write your Product Brief.

Imagine you’re launching yourself as a new offering. Document:

• What expertise am I selling? (Be specific—not “leadership” but “transforming legacy teams into AI-enabled organizations”)

• Who is my target customer? (What type of company, leader, or problem?)

• What outcomes do I deliver? (Measurable results, not activities)

• What makes me different? (Your unique combination of skills, experience, perspective)

• What is my pricing? (What are you worth? What should you be worth?)

• How am I taking this to market in 2026? (Your visibility strategy—network, content, positioning)

A word of warning: Don’t underestimate the time this takes. Finding the right words—the precise language that captures your unique value—requires iteration. Most people give up too early, settling for vague descriptions that sound like everyone else.

This is where AI becomes your thought partner. Use conversations with Claude or ChatGPT to pressure-test your positioning. Ask it to poke holes. Ask it to simplify. Ask it what’s unclear. The back-and-forth will sharpen your thinking faster than staring at a blank page.

Then pressure-test it with humans:

• Is this easy to understand in 30 seconds?

• Is it compelling enough that someone would forward it?

• Would you buy this?

• Does it reflect where the market is heading, not just where it’s been?

If the answer to any of these is no, you have work to do. That’s not a failure—that’s the beginning of clarity.

The Bottom Line

The professionals who thrive in the next decade won’t be the ones hoping their company takes care of them.

They’ll be the ones who stopped waiting for permission. Who stopped outsourcing their worth to an employer’s opinion. Who took ownership of their expertise, packaged it clearly, adapted it continuously, and made it impossible to ignore.

You are the product.

The only question is whether you’ll build one worth buying.

I’d love to hear from you: What’s your one-sentence value prop? Drop it in the comments—I’ll reply to as many as I can. Sometimes an outside perspective is exactly what sharpens the message.