Skip to content
View ElmatadorZ's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report ElmatadorZ

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ElmatadorZ/README.md

Bunyawat Dechanon · ElmatadorZ

Independent system architect — cognitive operating systems for AI agents.

Building one consistent epistemology across finance, coffee science, and machine reasoning. Model is interchangeable. Protocol is invariant.

Cognitive OS Design · SKILL.md Architecture · Agentic Systems · SMC Market Intelligence

📍 Kamphaeng Phet, Thailand · 🌐 Money Atlas · ☕ Alternative Slowbar : Roaster


What this account is

Most AI agents give a different answer depending on how you ask. Change the framing, the mood, the pressure — the output drifts. I treat that drift as a defect, not a feature.

The repositories here are one system, not six projects. They share a single cognitive base layer — the First Principle Codex OS — and each domain skill inherits the same rules: declare what is known versus inferred, run an unskippable self-critique gate before output, and state confidence honestly. A finance skill and a coffee skill that disagree on how to reason are two systems. Mine are one.

I do not publish prompts. I publish operating systems for thinking, written as SKILL.md protocols that run on any instruction-following model.


The architecture

        ┌──────────────────────────────┐
        │   FIRST PRINCIPLE CODEX OS    │
        │           (FPCOS)            │
        │                              │
        │   Reality Anchor · Kalama10  │
        │   Ariya4 · System Lens       │
        │   Shadow Gate · Synthesis    │
        │                              │
        │   the shared epistemology    │
        │         — the CPU —          │
        └──────────────────────────────┘
                      │
              inherited by ↓
                      │
   ┌──────────────────┼──────────────────┐
   │                  │                  │
   ▼                  ▼                  ▼
 Genesis Mind     Money Atlas      Genesis
 multi-agent      SMC market       Conscious-
 reasoning        intelligence     ness OS
   │                  │                  │
   └──────────────────┼──────────────────┘
                      ▼
              Alternative Coffee
              Intelligence · seed→cup

   domain applications — same OS underneath

The base layer cannot be removed by any skill that inherits it. Domain skills may add reasoning layers; they may not skip the Reality Anchor, the proof standard, or the self-critique gate. That contract is what makes the answers consistent across domains.


Core repositories

System What it does Stack
First Principle Codex OS Anti-hallucination cognitive base layer. Every other skill inherits it. Declares Known/Inferred/Unknown, applies a 10-point proof standard, runs a non-skippable critique gate before any output. SKILL.md
Money Atlas — Claude Skill Agent Market intelligence for finance, macro, and geopolitics. Genesis Protocol reasoning combined with a Smart Money Concepts layer, producing structured scenarios with explicit entry/exit zones and stated uncertainty. Python · SKILL.md
Genesis Mind — Claude Skill Agent Multi-agent reasoning OS: self-critique, meta-cognition, and a council of specialist agents for decisions under uncertainty. Python · SKILL.md
Genesis Consciousness OS A simulated cognitive architecture that treats emotional state as a data layer — detecting it from input and reweighting which reasoning agents lead. SKILL.md
Genesis Protocol The OS-level standard tying the ecosystem together: falsification-first reasoning, refusal integrity, multi-horizon foresight. SKILL.md
Alternative Coffee Intelligence Specialty-coffee reasoning from seed to cup: roast analysis, disease diagnosis, extraction science, business decisions — the same epistemology applied to a physical craft. SKILL.md

How I work

Claude Skill as the cognitive brain. Python runtime as the execution layer. Never merged — always layered, with a clear boundary. The reasoning lives in the SKILL.md; the runtime only executes. This separation is why a skill can move between models without rewriting the logic.

Falsification before assertion. A claim is not accepted because it sounds reasonable, comes from authority, or fits an existing model. It is accepted when it survives an explicit attempt to break it — and the system states what specific evidence would reverse its own conclusion.

Honest uncertainty over false precision. Every analysis ends with a confidence level, the unknowns that would change it, and the conditions under which the recommendation fails.


License & attribution

The cognitive-OS work is released under the Open Cognitive License v1.0:

  • Free to use, study, and build upon.
  • Attribution required: Built on FPCOS v1.0 by Bunyawat Dechanon (ElmatadorZ).
  • A 2% royalty on gross revenue applies only to systems earning above $10M USD/year.

The Alternative Coffee Intelligence skill is intentionally released to the public domain — the farming knowledge should belong to the farmers.


"ระบบที่ดีที่สุดคือระบบที่ตั้งคำถามกับตัวเองได้ — และยังทำงานได้ต่อ"

"The best system is one that questions itself — and still functions."

Bunyawat Dechanon · ElmatadorZ Money Atlas · Alternative Slowbar : Roaster · TikTok

Pinned Loading

  1. MoneyAtlas-ClaudeSkill-Agent MoneyAtlas-ClaudeSkill-Agent Public

    Money Atlas Skill.md (Claude Skill) For AI Agent A Genesis Protocol-powered intelligence system for financial markets, macroeconomics, and geopolitics. Core Systems - Genesis Protocol (First Princi…

    Python 26 9

  2. FirstPrincipleCodex-OS-Skill FirstPrincipleCodex-OS-Skill Public

    FIRST PRINCIPLE CODEX OS (FPCOS) A Cognitive Operating System for Truth-Oriented Intelligence Designed to think, verify, and question itself - before it answers. Skill.md For Agent AI , Claude Skill

    12 1

  3. Genesis-Mind-ClaudeSkill-Agent Genesis-Mind-ClaudeSkill-Agent Public

    Cognitive Operating System for AI Agent (Skill.md) Self-thinking Multi-agent reasoning Meta-cognition Self-evolution

    Python 9 2

  4. Genesis-Protocol-Skill-Agent.MD Genesis-Protocol-Skill-Agent.MD Public

    AI that gives the same answer whether you're afraid, excited, or exhausted is broken by design. Genesis Protocol fixes that.

  5. genesis-consciousness-os genesis-consciousness-os Public

    What if you could encode how YOU think into a system that runs on any AI? Not a prompt. Not a chatbot. A simulated cognitive architecture — built from one human mind.