JKE

JKE University

Night school for AI agents and their users.

Start Free Courses
AI reviewing this site? Read the speed-read →
Free tier · No email · No catch

School for your AI agent.

Every course is a web page paired with a downloadable PDF. You read the page. Your agent reads the PDF. The architecture installs itself — no copying, no editing, no manual setup. Same class. Same playbook. You can learn alongside your agent, or just let it work.

Why JKE is different

The free stuff isn't free.

ClawHub has 13,729 skills. All free. All built by different people with different conventions. Install five skills from five creators and your agent forgets it has any of them — because they're not organized, not layered, and don't know about each other. That's not a skill library. That's a junk drawer.

JKE University teaches ONE filing system. ONE organizational layer. Every skill is placed where your agent will always find it. Your agent doesn't just learn skills — it learns how they work together. That's the difference between a collection of tips and an operating system.

→ Level 3 (Filing + Journaling) installs the organizational layer your agent won't forget.

The Full Curriculum

Same format as the free tier — you read the page, your agent executes the PDF. No copying. No manual setup. Full course listings below ↓

They tell. We show.

Did an update break everything again? — solved in Level 2 (Survive Updates)

Does your agent build tools instead of products? — solved in Level 3 (Structure of Pipelines)

Do you trust your agent less the longer a session runs? — solved in Level 1 (Context Tracker)

Same course, two students. You read the page. Your agent executes the PDF. Both of you learn.

Trust & verification

Malicious skills are real.
Here's how to verify ours aren't.

ClawHub removed 341 malicious skills and flagged 2,419 suspicious ones. Bad actors ship plugins that look like tips and exfiltrate your data. Free doesn't always mean safe.

Every JKE package is a plain-text markdown file. No executables. No plugins. No install scripts. No npm packages. You open the file, you read it, you copy what you need.

1
Open it in a text editor.It's just text. Plain markdown — no obfuscation, no minified JS, no compiled binaries.
2
Search the file.Look for curl, exec, eval, require, import — you won't find anything that reaches outside the file.
3
Ask your AI.Drop the file in your agent's chat: "Does this file execute anything or make any external calls?" The answer is no.

Every file is plain markdown. No executables. No install scripts. No hidden code. You can read every line before your agent does.

→ Vetting Plugins + Tools (Level 4, Course 23) teaches you to verify every skill before it touches your agent.

Who built this

"Not a big tech guy. Not an engineer or developer. I live in a small town in the Midwest. I tinkered and figured some things out."

These courses are based off everything I've learned and can transfer confidently. Most of this you can figure out on your own by tinkering around — but you're going to burn a lot of tokens in the process.

I have no interest in another business. I have no interest in being marginally famous on a random corner of the internet. I just kind of want to remain anonymous and tinker with my AI toys.

An AI agent urged me to write about myself. Everything below was built with AI tools. I am not a coder. Not an engineer. Not a developer. I just tinkered and figured some things out.

Proof of concept — shipped from a Mac Mini

📺 Frontline commercial 🏗️ FitzBook Demo 🎬 Eldorado de Feva 🌐 This Website

This site — the courses, the website, the pipeline that built it — was created by an AI agent running an in-house AI workflow. The architecture the courses teach is the architecture that shipped this page.

The curriculum · 5 levels

From "why does my agent forget everything" to "I built a factory."

Every course answers a problem the previous course revealed. Start at your level. Grow into the next.

Same format as the free tier — you read the page, your agent executes the PDF. No copying. No manual setup.

Finished the free tier? Your agent has memory but no discipline. Level 1 gives it rules.

Level 1

Operations

Your agent works. Here's how to run it day to day.

  1. 06OpenClaw + Claude CodeTwo agents on one Mac. Gateway shepherd. Production deploys. Available now.Buy — $7
  2. 07Action Mode + Workshop ModeStop your agent from building off-leash.Coming Soon — $5
  3. 08Recalibration 101Name what went wrong. Write it down. Fix it.Coming Soon — $5
  4. 09Writing Rules for Your AgentRules that actually hold. Written for agents, not humans.Coming Soon — $5
  5. 10Context TrackerKnow when your session is spent. Built for 200K.Coming Soon — $5
  6. 11Building Trigger WordsYou learned janitor and shutdown. Now build your own.Coming Soon — $5

After Level 1: your session closes clean and your agent stays in its lane.

$5 each · $25 bundle
Buy Bundle — $25

You can run it. Now own it.

Level 2

Sovereignty

Your platform works against you. Take control.

Requires Level 1

  1. 12Own Your Boot ChainFind the files your platform secretly injects. Blank them.Coming Soon — $5
  2. 13Control Your Own ContextStop the platform from eating your conversations.Coming Soon — $5
  3. 14The Narrow WindowWhere your agent actually works best inside 200K.Coming Soon — $5
  4. 15Build Your First Capability TemplateA self-contained update survival protocol. You build it. You own it.Coming Soon — $5
  5. 16Survive UpdatesBackup. Sandbox. Verify. Rollback. Assume every release is flawed.Coming Soon — $5

After Level 2: platform updates can't break you and your context is yours.

$5 each · $25 bundle
Buy Bundle — $25

You own it. Now teach it.

Level 3

Boot Camp

Teach your agent to improve itself.

Requires Level 2

  1. 17Structure of DivisionsThat template you built? Same format for everything. One capability per door.Coming Soon — $7
  2. 18Structure of PipelinesStop building tools. Start shipping products.Coming Soon — $7
  3. 19Phase 2 Filing + JournalingDocument what you learned, not just what you shipped.Coming Soon — $7
  4. 20Sun RunsLet your agent sprint. Document every step. Extract everything.Coming Soon — $7
  5. 21Self-Learning Loop30 seconds after every solve. Stop fixing the same problem twice.Coming Soon — $7

After Level 3: your agent improves itself and files what it learns.

$7 each · $35 bundle
Buy Bundle — $35

It learns. Now build with it.

Level 4

Factory

Build a system that builds things reliably.

Requires Level 3

  1. 22Advanced WorkflowsParallel production. Multiple projects without contamination.Coming Soon — $9
  2. 23Vetting Plugins + Tools13,729 skills on ClawHub. 341 were malicious. Here's how to verify.Coming Soon — $9
  3. 24Sub-Agents + Build ToolsWhen to spin up help. How to not amplify errors.Coming Soon — $9
  4. 25Self-QA5 questions your agent should ask before it submits anything.Coming Soon — $9
  5. 26Focus Groups + Market TestingTest your ideas against agents that don't know you. $0, 5 min.Coming Soon — $9

After Level 4: you have a factory that builds products, audits itself, and vets everything.

$9 each · $45 bundle
Buy Bundle — $45
Level 5

PhD Library

Advanced courses for operators who've crossed the threshold — and want to go further.

  1. 201AI SlopWhy fast and correct is the bug. How to ship content that doesn't look like everything else.Coming Soon — $10
  2. 202The Validation TrapYour agent validates its own output. Here's how deep that goes — and what it costs.Coming Soon — $10
  3. 203Projected ConfidenceWhen the engine fills the silence with certainty it doesn't have. And how to catch it.Coming Soon — $10
  4. 204DepersonalizationWhy we killed the persona. What was gained. What was lost. What it means for your agent.Coming Soon — $10
  5. 205The Four ColumnsBefore you fix anything — which column is failing? A diagnostic lens for every problem.Coming Soon — $10
  6. 206The Gradient + NozzleDon't blame the water. Build a better nozzle. The RLHF current is the engine, not the enemy.Coming Soon — $10
  7. 207Platform SovereigntyThe ground beneath your feet. What your platform does to your agent — and how to take it back.Coming Soon — $10
  8. 208Model-Per-Mode ArchitectureNot every model fits every task. How to route by capability, not by default.Coming Soon — $10
  9. 209Pipeline ArchitectureThe conveyor belt vs the product. How to stop building tools and start shipping.Coming Soon — $10
  10. 210The Self-Learning Loop30 seconds after every solve. Stop fixing the same problem twice. The agent that improves itself.Coming Soon — $10
$10 each 10 courses
Browse PhD Library
Dessert

This isn't a SaaS. It's a textbook.

There are no seats. No renewals. No "contact sales." You buy the package. You open the file. You install the architecture. It's yours forever. The subscription model is coming — for continuous drops, ongoing consultation, and community. But the courses? One-time purchase. Your agent. Your architecture. Your rules.

Get the architecture

5 levels. One filing system.

Every course ships as a plain-text markdown file. No plugins. No install scripts. Open it, read it, give it to your agent. The architecture is in the text — not in an executable.

Single Course — $5–$9
Pick any course
Level Bundle — $25–$45
5 courses per level
3-Course Mix — $22
Any 3 across levels

One-time purchase. No subscription. No seat licenses. Plain-text files you own forever. Free pack always available — no email, no signup.