foundationPrime Your State Before AI
Execute a cold start ritual before opening any AI interface.
⏱ 5m⚡ low
Neuro-Cybernetic Interface
Protocol
The Agentic Operator
You are not a button-pusher. You are the architect of semi-autonomous cognitive systems. These 20 protocols keep your brain sharp, your agents aligned, and your output pristine.
Old Way
- Offload everything to AI for maximum speed
- Trust AI outputs because they look correct
- Let agents run autonomously without limits
- Use default settings for all tasks
- Close chat windows without extracting value
New Way
- Protect germane cognitive load to prevent atrophy
- Verify reasoning chains before accepting conclusions
- Enforce hard-coded limits and HITL gates
- Tune temperature and context per task type
- Debrief every session into your knowledge system
Myth
Common Myth
AI makes you more productive by doing the thinking for you.
Mechanism
The Five Control Systems
Click a system to explore its habits
Neuro-Cybernetic Interface
Maintain the biological hardware. Counteract cognitive atrophy through deliberate friction and dopamine regulation.
Agentic Governance
Command and control for autonomous loops. Kill switches, permission scoping, and human-in-the-loop gates.
Information Logistics
Context engineering and knowledge management. Garbage collection, PARA structure, and prompt version control.
Verification & Truth
Counteract hallucination and drift. Chain of thought injection, red teaming, and code review checklists.
Operational Velocity
The temporal interface. Timeboxing, session debriefs, and sovereignty checks before shipping.
Getting Started
Step01
Week 1: Establish the Foundation
Set up hard-coded agent limits, configure kill switches, and start priming your state before AI sessions. Build the safety infrastructure.
Step02
Week 2: Build Governance Muscle
Run your first kill switch drill, scope agent permissions, and implement context garbage collection. Start the weekly Zero-AI blocks.
Step03
Week 3: Sharpen Verification
Demand Chain of Thought on all complex queries. Run your first Manual Audit Lottery. Click every citation.
Step04
Week 4: Optimize Flow
Separate Maker and Manager blocks. Implement session debriefs. Run the Signature Test before every ship.
Step05
Ongoing: Master the Loop
Calibrate limbic friction, randomize rewards, red team strategic decisions, and inject 10% human signal into every output.
A Day in Your Life
Morning Prime
06:00View sunlight for 10-30 minutes. Execute cold start ritual before opening any AI interface. Set dopamine baseline.
Deep Work Block
08:00Maker mode. Wait 10 minutes before AI. Sketch solutions first. Timebox prompt refinement to 20 minutes max.
Agent Supervision
12:00Manager mode. Review agent outputs. Check token burn rate. Run HITL approvals for queued actions.
Zero-AI Block
14:0060 minutes of cognitive resistance. No AI assistance. Perform core competency tasks manually.
Session Debrief
17:00Extract key decisions to PARA system. Note one AI failure. Run Signature Test on deliverables. Close all loops.
Cognitive Resistance
3x60min/week
Manual Audit Rate
10%
Max Agent Iterations
10loops
Context GC Threshold
70%
Human Signal Injection
10% min
Prompt Timebox
20min max
Progression
Habit Tiers
Tier 1
Foundation
Safety Infrastructure — Set hard-coded limits. Configure kill switches. Prime state before AI. Segment topics per chat. Demand Chain of Thought.
Tier 2
Growth
Active Governance — Drill kill switches weekly. Scope agent permissions. Run manual audit lottery. Maintain PARA structure for RAG.
Tier 3
Mastery
Adversarial Verification — Red team your AI. Cross-reference with second models. Randomize HITL checks. Refactor until you can explain every line.
Tier 4Peak
Titan
Sovereign Mastery — Audit cognitive load weekly. Run code security checklists. Inject 10% human signal. You are the architect, not the tool.
"The future belongs not to those who use AI the most, but to those who govern it the best.
Frequently Asked Questions
Protocol Playbooks
Curated sequences of habits designed to be practiced together. Click a playbook to see its cards in the deck below.
4 cards
Morning Prime Protocol
Set your neurochemical baseline and enter AI sessions in a directive state.
Prime Your State Before AI
View Morning Sunlight
Wait 10 Minutes Before AI
+1 more habits
7 cards
Deep Work Block
4 hours of high-output work with proper AI governance.
Wait 10 Minutes Before AI
Timebox Prompt Refinement
Separate Maker and Manager Blocks
+4 more habits
5 cards
Agent Deployment Checklist
Safety infrastructure before running any autonomous agent.
Set Hard-Coded Limits
Configure Kill Switch
Scope Agent Permissions
+2 more habits
5 cards
Weekly Governance Review
Maintain system integrity and prevent drift.
Drill the Kill Switch
Audit Your Cognitive Load
Maintain PARA Structure
+2 more habits
7 cards
Pre-Ship Verification
Quality assurance before any deliverable leaves your desk.
Run Manual Audit Lottery
Click Every Citation
Red Team Your AI
+4 more habits
Quests
Challenges to accelerate your transformation. Click a quest to see its target cards.
Initiation
Sovereign Initialization
Establish a neuro-cybernetic baseline by priming your state, setting constraints, and demanding reasoning before AI interaction.
"The grid is loud. Silence it. Reclaim your biological directive before the silicon takes over."
FOUNDATION tier streak for 7 days
6 cards
View Quest Cards
System Mastery
The Governor's Mandate
Secure the operational perimeter by mastering kill switches, API permissions, and human-in-the-loop oversight gates.
"An agent without a leash is a liability. Command the machine; do not be its passenger."
Complete Agentic Governance system
8 cards
View Quest Cards
Titan Trial
The Proof of Humanity
Reclaim the final output by injecting 10% human signal and passing the rigorous Signature Test.
"The machine can mimic, but it cannot sign. Put your soul back into the output."
Complete TITAN tier
2 cards
View Quest Cards
The Full Deck
40 habits across 5 core systems
Systems:
Playbooks:
foundation
foundationView Morning Sunlight
Get 10-30 minutes of sunlight within an hour of waking to set dopamine baseline.
⏱ 15m⚡ low
Neuro-Cybernetic Interface
growthConduct Zero-AI Blocks
Schedule three 60-minute analog sessions weekly to maintain core competencies.
⏱ 1h⚡ high
Neuro-Cybernetic Interface
growthRun Comparative Analysis
After analog work, run the same task through AI and diff the outputs.
⏱ 15m⚡ medium
Neuro-Cybernetic Interface
masteryCalibrate Limbic Friction
Increase task complexity when analog work feels effortless to maintain growth.
⏱ N/A⚡ high
Neuro-Cybernetic Interface
masteryRandomize Your Rewards
Use stochastic celebration—flip a coin to decide whether to reward milestone completion.
⏱ 1m⚡ low
Neuro-Cybernetic Interface
titanAudit Your Cognitive Load
Categorize workflows into Grunt/Tactical/Germane buckets and protect germane load.
⏱ 30m⚡ medium
Neuro-Cybernetic Interface
foundationSet Hard-Coded Limits
Configure MAX_ITERATIONS and BUDGET_CAP at the API wrapper level for all agents.
⏱ 15m⚡ medium
Agentic Governance
foundationConfigure Kill Switch
Set up a single-command termination signal for instant agent shutdown.
⏱ 10m⚡ low
Agentic Governance
growthDrill the Kill Switch
Execute a weekly fire drill to test agent termination latency and build muscle memory.
⏱ 5m⚡ low
Agentic Governance
growthScope Agent Permissions
Create tiered API keys with Read-Only, Draft, and Executive access levels.
⏱ 30m⚡ medium
Agentic Governance
masteryImplement HITL Gates
Configure agents to pause and request explicit approval before irreversible actions.
⏱ 20m⚡ medium
Agentic Governance
masteryInject Random HITL Checks
Stochastically interject approval gates in low-stakes tasks to maintain vigilance.
⏱ N/A⚡ low
Agentic Governance
titanMonitor Token Burn Rate
Check daily token spend against 30-day average and set 150% anomaly alerts.
⏱ 5m⚡ low
Agentic Governance
titanAudit Post-Kill Traces
After any kill switch activation, dump and analyze the agent's Chain of Thought log.
⏱ 15m⚡ medium
Agentic Governance
foundationSegment Topics Per Chat
Maintain one goal per chat thread to prevent semantic bleed and hallucinations.
⏱ N/A⚡ low
Information Logistics
foundationPrime New Sessions
Start every chat with a Role/Context/Constraints primer using KERNEL or COSTAR.
⏱ 2m⚡ low
Information Logistics
growthRun Context Garbage Collection
At 70% context window usage, summarize progress and reset to a fresh session.
⏱ 5m⚡ low
Information Logistics
growthMaintain PARA Structure
Organize all files into Projects/Areas/Resources/Archives for machine-readable RAG.
⏱ 30m⚡ medium
Information Logistics
growthCreate Context Files
Add a context.md to every project root with Goal, Status, Constraints, and Style Guide.
⏱ 10m⚡ low
Information Logistics
masteryVersion Control Prompts
Maintain a Git repository for master prompts with changelogs and A/B test results.
⏱ 10m⚡ medium
Information Logistics
masteryModularize Prompt Components
Break monolithic prompts into Role/Format/Constraint modules for dynamic assembly.
⏱ 30m⚡ high
Information Logistics
titanTune Temperature Per Task
Set temperature 0.0-0.2 for code, 0.3-0.7 for drafts, 0.8+ for brainstorming.
⏱ 1m⚡ low
Information Logistics
titanRun Double-Pass Generation
Generate ideas at high temp, then refine at low temp for feasibility and logic.
⏱ 15m⚡ medium
Information Logistics
foundationDemand Chain of Thought
Never accept naked answers—require step-by-step reasoning before conclusions.
⏱ N/A⚡ low
Verification & Truth
foundationRead Reasoning First
Evaluate the AI's logic chain before looking at its final answer.
⏱ 2m⚡ low
Verification & Truth
growthRun Manual Audit Lottery
Randomly select 10% of AI outputs for deep, line-by-line verification.
⏱ 15m⚡ medium
Verification & Truth
growthClick Every Citation
Verify that AI-provided URLs and references actually exist and say what's claimed.
⏱ 5m⚡ low
Verification & Truth
masteryRed Team Your AI
Challenge recommendations aggressively and demand sources for strategic claims.
⏱ 10m⚡ medium
Verification & Truth
masteryCross-Reference With Second Model
Verify critical factual claims by checking against a different LLM or search.
⏱ 5m⚡ low
Verification & Truth
titanRun Code Security Checklist
Check AI code for input sanitization, edge cases, and hallucinated dependencies.
⏱ 10m⚡ medium
Verification & Truth
titanRefactor to Understand
Manually refactor AI code until you can explain every line—delete what you can't.
⏱ 15m⚡ high
Verification & Truth
foundationWait 10 Minutes Before AI
Sketch your solution manually before prompting to ensure you understand the problem.
⏱ 10m⚡ medium
Operational Velocity
foundationTimebox Prompt Refinement
Limit prompt tweaking to 20 minutes—if it doesn't work, do the task manually.
⏱ 20m⚡ medium
Operational Velocity
growthSeparate Maker and Manager Blocks
Dedicate distinct time blocks for deep work versus AI supervision and prompt tweaking.
⏱ N/A⚡ low
Operational Velocity
growthDebrief Every Session
Before closing a chat, extract key decisions and lessons into your knowledge system.
⏱ 5m⚡ low
Operational Velocity
masteryAnchor Reward to Orchestration
Attach dopamine to your prompt architecture, not the AI's output.
⏱ N/A⚡ low
Operational Velocity
masteryUse Visual Timers
Keep a visible countdown during AI sessions to maintain urgency and time awareness.
⏱ N/A⚡ low
Operational Velocity
titanRun the Signature Test
Before shipping, ask: 'Would I be proud to sign this if AI didn't exist?'
⏱ 2m⚡ low
Operational Velocity
titanInject 10% Human Signal
Ensure at least 10% of every output is purely yours—voice, anecdote, or novel insight.
⏱ 10m⚡ medium
Operational Velocity
Sources & References
External reading that informed this stack.
- 01
The Google Effect: How Search Engines Change Memory
Science Magazine
science.org
- 02
Cognitive Load Theory: Implications for Learning
Educational Psychology Review
link.springer.com
- 03
Dopamine, Motivation, and the Neuroscience of Reward
Huberman Lab
hubermanlab.com
- 04
Human Factors in Automation: Vigilance and Complacency
Human Factors Journal
journals.sagepub.com
- 05
The PARA Method: A Universal System for Organizing Digital Information
Forte Labs
fortelabs.com
- 06
Prompt Engineering Best Practices
OpenAI Documentation
platform.openai.com
- 07
Lost in the Middle: How Language Models Use Long Contexts
arXiv
arxiv.org
- 08
AI Security: Non-Human Identity Management
OWASP
owasp.org
Explore More Identity Stacks
33 cards6 systems
The Performance Stacker
Build physical capacity as your ultimate work multiplier. A habit system for WFH entrepreneurs who refuse to let fitness be an afterthought.
Movement SnacksAnchor WorkoutsProgress Tracking+3
Performance Protocol
30 cards5 systems
The Bootstrapped Industrialist
A systems-level protocol for building a $1M+ business with zero employees. Leverage code, media, and automation to replace labor.
Shipping VelocityAutomated OperationsMargin Fortress+2
The Machine
30 cards5 systems
The Workspace Orchestrator
Eliminate the cognitive tax of 'Window Soup' through spatial determinism and proprioceptive navigation.
Spatial Memory OrchestrationProprioceptive Input OptimizationZero-Clutter Cognitive Hygiene+2
The Architect