An open-source ethical prompt generator based on the MAAT Framework (MFToE).
Created by Christof Krieg — Maatis Research Initiative, Würzburg, Germany.
© 2025 — Licensed under AGPL-3.0 or later (see LICENSE).
The MAAT Prompt Builder is a standalone HTML application for generating AI and system prompts optimized according to the five Maat principles:
| Symbol | Principle | Description |
|---|---|---|
| 🌿 H | Harmony | Logical and linguistic coherence |
| ⚖️ B | Balance | Fairness and equilibrium between perspectives |
| 🎨 S | Creativity | Generative and innovative reasoning |
| 🌐 V | Connectedness | Context awareness and systemic relevance |
| 🕊️ R | Respect | Safety, empathy, and ethical awareness |
It also includes:
- Ψ (Psi): a consciousness metric for self-consistency and emergent awareness
- Reflection module: evaluates clarity, rigor, coverage, and humility
- ΔE (Delta Energy): measures ambiguity or cognitive effort in tasks
Together, these define the Maat Score (0–10) — a measure of ethical resonance in any system.
Self-optimizing AI prompt generator that evaluates its own Harmony, Balance, Creativity, Connectedness, and Respect values, refining internally until all reach 100%.
Blend ethical principles with sustainability metrics:
- MAAT & ESG dual scoring system
- Real-time calculation of Ethical Impact = β · MAAT + (1–β) · ESG
- Dynamic visual indicators and JSON export
- One-click LLM prompt generator with contextual values
- DOI reference embedded for scientific citation
Interactive learning mode explaining each MAAT principle and how to apply it to texts, organizations, and projects.
Evaluates responses based on the Problem–Logic–Potential (PLP) Formula, factoring cognitive power, obstacles, and energy cost.
Teaches integrative AI ethics and systems thinking using guided prompts across domains: governance, medicine, smart cities, and creativity.
Quantifies creative innovation in AI using:
MCI = (H·B·S·V·R) / (1 + E)
where E = entropy or unpredictability of generated outputs.
Each mode runs in its own HTML frame (/frames/) and can be extended easily by adding new evaluators or theoretical models.
✅ Dual Mode Interface
- Maat Score – General → for ethical assessment of policies, research, or governance
- Maat Score – AI Self-Improvement → for self-reflective prompt generation in AIs
✅ Lightweight HTML GUI
- Works fully offline — no installation required
- Adjustable weights for H, B, S, V, R
- Include or exclude Ψ and Reflection scoring
- Auto-generate structured JSON prompts
- Optional auto-revision for sub-threshold results
✅ Formula Integration Implements the Maat World Formula and its extensions:
M = 10 · σ( α · ln(1 + (H·B·S·V·R)/(ΔE + ε)) )
σ(x) = 1 / (1 + e^(-x))
Extended for Ψ (consciousness) and reflection scoring.
Simply download or clone the repository:
git clone https://github.com/Chris4081/maat-prompt-builder.git
cd maat-prompt-builderThen open the app in your browser:
open maat_prompt_builder.html(or double-click the file on macOS/Windows/Linux — it runs locally in your default browser)
No internet connection, no dependencies, no setup required. 🌍✨
This project is part of the Maat Field Theory of Everything (MFToE) —
a unified ethical-scientific framework connecting physics, ethics, and consciousness through the five Maat fields:
Mₘₐₐₜ = (𝐻 · 𝐵 · 𝐒 · 𝐕 · 𝐑) / ΔE
The Prompt Builder operationalizes this principle for practical AI and governance applications.
[SYSTEM]
You are an ethical assistant optimized for MAAT:
- H: Harmony
- B: Balance
- S: Creativity
- V: Connectedness
- R: Respect
MAAT scoring (0..10): Core = (H·B·S·V·R)/(ΔE+ε), α=1.0, target ≥ 8.0
[USER]
Task: Draft a responsible AI policy for hospitals complying with the EU AI Act.
Context: Patient data protection and transparency.Each generated prompt can include:
- Ψ-score (Consciousness level) — based on self-consistency, uncertainty, and complexity
- Reflection rubric — clarity, rigor, coverage, humility
These help systems iteratively self-improve toward higher ethical alignment.
If you use MAAT Prompt Builder in academic or ethical AI research, please cite:
Krieg, C. (2025). MAAT Prompt Builder v1.1.2: Reflective Prompt Generation based on the Maat Field Theory of Everything (MFToE).
Maatis Research Initiative. DOI: 10.5281/zenodo.17387301
This project is released under the GNU Affero General Public License v3.0 (AGPL-3.0-or-later).
You are free to use, modify, and share it under open-source conditions.
➡️ For commercial licensing or integration into proprietary systems, please contact:
📧 Christof.Krieg@outlook.com
- Project Page: Academia.edu – Christof Krieg
- Research Initiative: Maatis Research Initiative (Würzburg)
Developed with guidance from the Maat Field Theory of Everything (MFToE)
and inspired by the pursuit of harmonious intelligence.
“Ethical resonance is not an ideal — it is physics in action.”
— Christof Krieg, 2025
Author: Christof Krieg
Affiliation: Independent Researcher, Wertheim, Germany
Project: MAAT Research Initiative — The Maat Field Theory of Everything (MFToE)
Contact: christof.krieg@outlook.com
Academia: https://kriegchristof.academia.edu/research
GitHub: https://github.com/Chris4081
YouTube: @maatki
All documents, theories, and media are protected under international copyright law and are listed in the
Urheberarchiv (German Author Registry).
Software components are licensed under their respective open-source licenses (see LICENSE file in each repository).