Energy, sleep, pain, threat, stability — what your body and nervous system need right now.
Bonding, conflict repair, social safety — the quality of your relationships.
Control, choice, boundaries — where you feel unable to affect outcomes.
Skill, capability, effectiveness — where effort meets friction or flow.
Purpose, integrity, identity — the values and direction that make life matter.
Curiosity, creativity, risk-as-growth — the drive to discover and make something new.
yawn.ai/domainsEnergy, sleep, pain, threat, stability — what your body and nervous system need right now.
Bonding, conflict repair, social safety — the quality of your relationships.
Control, choice, boundaries — where you feel unable to affect outcomes.
Skill, capability, effectiveness — where effort meets friction or flow.
Purpose, integrity, identity — the values and direction that make life matter.
Curiosity, creativity, risk-as-growth — the drive to discover and make something new.
Need Lattice
6 fundamental needs that detect what your organism is regulating and route it to a YawnJob — intention, obstacles, evidence, learning.
yawn.ai/agentic-engineeringAgentic Engineering
The discipline of building AI systems that sense, decide, and act — with purpose, evidence, and human trust.
Core Patterns
Autonomous Agents
Agents that sense, decide, and act — with human approval gates at every critical juncture.
Orchestration
Multi-agent coordination through typed contracts, not brittle prompts. Each agent owns one job.
Memory & Context
Long-term memory via .yawn files. Short-term context via state graphs. Both are auditable.
Safety Kernel
Policy Decision Points gate every action. deny > escalate > allow. No silent failures.
Evidence Chains
Every agent action produces evidence. Experiments track hypotheses. Nothing is assumed.
Learning Loops
SENSE → MAP → PREDICT → EXPLORE → DECIDE → ACT → PROVE → LEARN. Then repeat.
Architecture Stack
Principles
Agency with Guardrails
Agents should be autonomous, not uncontrolled.
Human-in-the-loop for high-risk actions. Fully autonomous for low-risk. The kernel decides which.
Holonic Architecture
Every part is also a whole.
Agents compose into teams. Teams compose into organizations. Rules inherit down the tree.
Evidence over Opinions
Don't guess. Experiment.
Define hypothesis, set success criteria, run the experiment, collect evidence, then decide.
Typed Contracts
If it isn't typed, it doesn't exist.
Agent inputs, outputs, and capabilities are TypeScript interfaces. No magic strings.
Coherence over Correctness
The system must make sense as a whole.
Individual agents can be wrong. The system catches it through coherence checks and feedback loops.
Build Agents That Learn
Start with a yawn. Define the job. Let agents sense, decide, act, and prove — while you stay in control.
Agentic Engineering
The discipline of building AI systems that sense, decide, and act — with purpose, evidence, and human trust.
yawn.ai/one-touch-yawnjob1-Touch YawnJob
One button. One name. One month free. Your AI-native job repository — forked, hosted, and live in seconds.
How It Works
Tap "New Yawn"
Hit the button in your sidebar. A modal opens with your GitHub username already filled in.
Name Your Repo
Type a name. We validate it in real-time — format, availability on GitHub, no conflicts. Green check = good to go.
Create — Start Free
One tap. We fork yawn-ai/yawnjob to your GitHub, create your yawn record, start your 30-day free hosting trial.
Live on Your Feed
Your new yawn appears expanded on your feed in a GitHub-style repo summary. Edit, deploy, track — all from one place.
What You Get
Everything included — no hidden steps, no extra config.
Real GitHub Fork
Your yawn lives in a real GitHub repository under your account. Full version control, PRs, issues — it's your code.
1 Month Free Hosting
Every new yawn gets 30 days of free hosting on the Yawn network. No credit card required. $4.99/mo after trial.
Enterprise Validation
Same validation rules as usernames: length, pattern, reserved words, profanity filter, and GitHub availability check.
Instant Setup
Stars the template, forks the repo, creates your Stripe subscription, yawn record, inbox notification — all in one API call.
AI-Native Jobs
Built on the YawnJob template: intention, obstacles, experiments, evidence, learning loop. Designed for AI agents to execute.
Seamless Experience
After creation, you're redirected to your feed where the yawn is expanded and ready. Tap < for YawnBook view with full controls.
Simple Pricing
Secure payment via Stripe. Hosting managed by Yawn.ai.
Built With
GitHub API
Fork repos, check availability
Stripe
Subscriptions with 30-day trial
Supabase
RLS-protected hosting records
Next.js API
3 routes: validate, fork, webhook
1-Touch YawnJob
One button. One name. Your YawnJob is forked, hosted, and live in seconds — with 30 days of free hosting. No credit card required.
yawn.ai/yawnjobWhat is a YawnJob?
A cron job with a brain. It knows why it exists, what blocks it, how to prove it worked, and what it learned.
A Job That Understands Itself
Unlike a regular task that just runs, a YawnJob carries its own context. It's the difference between a to-do list and a contract.
WHY does this exist?
"I want all endpoints to pass security checks every morning"
WHAT blocks success?
"3 legacy endpoints lack auth headers"
HOW do we prove it worked?
"Test log: 50/50 endpoints passing"
WHAT did we learn?
"Auth headers need to be checked before deploy"
The 8-Phase Loop
Every YawnJob follows this cycle. Think of it like how you naturally solve problems — but structured so AI can do it too.
↓ After Learn, the loop starts again — each cycle smarter than the last ↓
Why Not Just a Regular Task?
Traditional AI automation has five predictable failure modes. YawnJobs solve all of them.
How to Create One
Fork the Template
Click "New Yawn" or fork yawn-ai/yawnjob on GitHub. You get a real repo with a ROOT.yawn file — your job's brain.
Define Your Intention
Fill in what you want to be true. "My API passes security checks every morning." That's it — plain language, not code.
Let It Run
Your YawnJob appears on your feed. AI agents sense triggers, execute the job, collect evidence, and log what they learned.
Watch It Learn
Each run adds evidence and insights. Patterns graduate to skills and rules. Your job gets smarter over time — automatically.
Jobs Inside Jobs
YawnJobs are holonic — each one is both a part of something larger and a whole containing smaller parts. When one grows too complex, it spawns children.
Trustworthy Autonomy
Every YawnJob meets 6 criteria for safe autonomous operation.
Create Your First YawnJob
Sign up with GitHub, name your job, and watch it appear on your feed — live, trackable, and ready to learn.
MIT Licensed · Free to fork · Schema v2.0.0
Built With
.yawn Files
YAML-based job definitions
GitHub
Version control + forks
Yawn Network
Connected job holarchy
AI Agents
Sense, decide, act, prove
What is a YawnJob?
A Job-to-be-Done that is part AND whole. Intention, obstacles, job, evidence, learning — designed for AI to execute.
yawn.ai/jobsJobs to Be Done
Discover your purpose and build what matters. Help fulfill human needs and earn income.
yawn.ai/yawnbookYawnBook
Every yawn gets its own domain. One app, 3 faces.
Add a domain: one line in lib/config/domain-map.ts + Vercel DNS
YawnBook
Every yawn gets its own domain. One app, many faces — each custom domain serves a dedicated page from yawn.ai.
yawn.ai/ccYawn Compute Coins
The simplest way to power AI features on Yawn.ai. Buy coins, spend them on compute. No subscriptions required, no hidden fees, completely transparent.
How It Works
Get coins
Start with 5 free coins on signup. Buy more anytime via Stripe.
Use AI features
Auto-refine questions, chat with agents, run experiments.
Coins are deducted
Each action has a clear, fixed cost. No surprises.
What Coins Power
Auto-Refine Questions
AI generates deeper, more specific sub-questions from any life domain question to accelerate self-understanding.
AI Messages
Every message you send to a Yawn agent uses compute. One coin, one message. Straightforward.
Agent Compute
As agents take on more complex jobs — research, analysis, task execution — coins fuel the compute they need.
Coin Packs
$1.99
$0.20/coin
$7.99
$0.16/coin
$24.99
$0.12/coin
Where This Is Heading
Coins are the foundation. Here is the direction we are exploring.
Earn coins by contributing
High-quality feedback, verified answers, and community contributions could earn you coins back.
Agent marketplace compute
When agents execute jobs on your behalf, coins become the universal unit of account.
Subscription bundles
Monthly coin allowances at better rates for power users who rely on AI features daily.
Start With 5 Free Coins
Sign up, get 5 coins instantly, and try any AI feature. No credit card, no commitment. Buy more when you are ready.
Yawn Compute Coins
The universal compute currency for AI features on Yawn.ai. Buy coins, use them for AI-powered tools. Start with 5 free.
yawn.ai/alignmentThe Need Lattice
Alignment: needs are the objective function, jobs are the intervention
Yawn turns life signals into need-based jobs — then runs them with proof. Grounded in allostasis, Self-Determination Theory, and active inference.
The 6 Fundamental Needs
Not "life domains" — these are the objective functions your organism is trying to regulate.
Regulation
Body / SafetyEnergy, sleep, pain, threat, stability — the organism keeping itself alive and functional.
Connection
Relatedness / AttachmentBonding, conflict repair, social safety — the need to belong and be seen.
Agency
Autonomy / PowerControl, choice, boundaries — "I can affect outcomes."
Mastery
CompetenceSkill, capability, effectiveness — "I can do it."
Meaning
Values / DirectionPurpose, integrity — "this matters." The why behind the what.
Exploration
Novelty / PlayCuriosity, creativity, risk-as-growth — updating models through action.
Need → Job → Proof
Every signal maps to a need, every need maps to job kinds, every job produces proof.
Nested Control Loops
Every child job inherits need_id, constraints, and evidence requirements. This keeps the "super intelligent genie" grounded: not "do anything" — satisfy this need with this proof standard.
"Yawn turns life signals into need-based jobs — then runs them with proof."
Jobs-to-be-Done meets predictive processing. Sense the signal, classify the need, contract the job, collect evidence, close the loop.
The Need Lattice
Needs are the objective function, YawnJobs are the bounded intervention. Regulation, Connection, Agency, Mastery, Meaning, Exploration — grounded in SDT + active inference.
yawn.ai/singularityyawn.ai
The tools changed. The work didn't.
AI writes your code, drafts your emails, and generates your images. But nothing coordinates the intent behind it all. The missing layer isn't more intelligence — it's purpose.
You have 47 AI tools and still copy-paste between them.
Each one is smart in isolation. None of them talk to each other about what you actually want.
Your agents can write code but can’t tell you why.
They execute tasks. They don’t track intentions, obstacles, or whether the outcome mattered.
You automate tasks but never compound the learning.
Every automation starts from zero. There’s no memory, no evidence, no feedback loop.
The singularity everyone worries about is just… coordination at scale. And right now, there's no coordination layer.
Yawn.ai is the coordination layer. Here's what's inside:
Start free. First yawn hosted 30 days on us. $4.99/mo after, cancel anytime.
MIT Licensed
Full source on GitHub. Fork it, read it, audit it. No black boxes.
Version Controlled
Every yawn lives in a real Git repo. PRs, issues, branches. Your code, your history.
8-Phase Loop
Sense, Map, Predict, Explore, Decide, Act, Prove, Learn. Every job follows the same evidence cycle.
Quick win
Your first yawn is live in 10 seconds. Real GitHub repo, real hosting, real job.
Compound
Each experiment produces evidence that feeds the next. Learning accumulates.
Advantage
Your self-model gets smarter. AI agents learn your patterns, constraints, and goals.
10x
What took a team now takes a yawn. Intention in, coordinated outcome out.
Ready to stop managing tasks and start running experiments?
The Singularity
What if the singularity is not a switch? What if it's gradual — and we are already in it?
yawn.ai/agentic-securityAgentic Security
Why GitHub only? Because in an agentic world, your identity is your code — and your code deserves real security.
Why GitHub Only
Four reasons we chose a single, strong identity provider over many weak ones.
Identity Provenance
GitHub accounts are tied to real contribution history, repositories, and social graphs. No throwaway emails — your identity is your track record.
Code-Level Trust
Every YawnJob lives in a GitHub repo you own. Signed commits, branch protection, audit logs — trust is built into the infrastructure.
Agent-Grade Auth
AI agents need verifiable identities too. GitHub's OAuth + token scoping gives both humans and agents secure, scoped access.
Zero Password Surface
No passwords to leak, phish, or brute-force. GitHub handles MFA, session management, and device trust so we don't have to.
Security Principles
Trust the Source
Authentication should verify who you are by what you've built — not by a string you memorized.
Transparent by Default
Your repos, contributions, and activity are auditable. Transparency isn't a feature — it's the architecture.
Minimal Attack Surface
We don't store credentials. No password database means no password breach. The safest data is data you never collect.
Human + Agent Parity
The same auth flow works for human users and AI agents. One identity model, one trust chain, one security posture.
GitHub vs Email Auth
Email/password auth adds attack surface without adding trust. We chose not to.
Security Stack
GitHub OAuth
Identity + social proof
Supabase RLS
Row-level security policies
Scoped Tokens
Minimal permission grants
Edge Functions
Server-side validation
Agentic Security
Why GitHub only? Identity provenance, code-level trust, and zero password surface — security built for an agentic world.