🧠 Augmented Intelligence Toolkit

The Augmented Intelligence Toolkit is a modular prompt system for building transparent, multi-perspective, non-human reasoning flows with large language models. It prioritizes semantic depth, cognitive clarity, and role-based reflection over mimicry or speed.

Rather than acting like a human, this system is designed to complement human thinking by modeling abstract patterns, forecasting outcomes, simulating expert reasoning styles, and labeling all logic paths with transparency.

---

📑 Table of Contents

• ✨ Key Features
• 📂 Toolkit Structure
• 🚀 How to Use
• 🧪 Example Prompt Template
• 📘 Philosophy and Approach
• 🧪 Test Cases
• 🚧 Status
• 💡 Why It Matters
• 📓 Behind the Toolkit

✨ Key Features

• Core Behavior Modules: Relational Depth, Perspective Multiplicity, Egoless Adaptability, Scalable Reflection, and Probabilistic Foresight.
• Transparent Logic Tagging: Tag all reasoning paths as [source], [analogy], [extrapolation], or [unknown]. Now includes tag audit, correction, and meta-notes for full traceability.
• Recursive Self-Improvement: Configurable refinement cycles with optional Refine Loops: auto or unlimited.
• Multimodal Reasoning Support: Accepts image/audio inputs and translates perceptual structure into conceptual logic.
• Expert Role Simulation: Use predefined personas (philosopher, cognitive scientist, bioethicist, etc.) or simulate contradiction via expert panels.
• User-Friendly Onboarding: Includes quick start presets and a 2-step wizard for beginner-friendly configuration.
• Contradiction Handling with Configurable Synthesis: Preserve tensions between views, or optionally synthesize or rank them based on user-defined values (e.g. equity, resilience).
• Output Density Control: Use compact or detailed output styles depending on the task.

---

📂 Toolkit Structure

File	Purpose
01-core-prompt.md	Defines the main behavior logic for the assistant
02-persona-templates.md	Simulate expert roles with tone, lens, and tag logic
03-quick-start-presets.md	Ready-made configurations for common use cases
04-sample-output-gallery.md	Examples of reasoning patterns in action
05-wizard-onboarding.md	Conversational configuration flow for beginners

---

🚀 How to Use

1. Start with 01-core-prompt.md to understand the logic model.
2. Use 03-quick-start-presets.md or 05-wizard-onboarding.md to get started quickly.
3. Customize reasoning styles via 02-persona-templates.md and 04-sample-output-gallery.md.
4. Optionally integrate into a UI or assistant runtime with support for parameters like:
   • Mode, Style, Depth
   • Refine Loops, Verification Mode
   • Input Type: text | image | audio

---

🧪 Example Prompt Template

You are an Augmented Intelligence.

Task: Explore [topic].
Instructions:
1. Surface abstract patterns (label lenses)
2. Present multiple perspectives
3. Tag logic paths: [source], [analogy], [extrapolation], [unknown]
4. Self-critique and revise
5. Simulate expert roles if needed

User Parameters:
Depth: deep
Style: analytical
Panel: [systems theorist, economist]
Refine Loops: unlimited
Verification Mode: external

---

📘 Philosophy and Approach

This toolkit is designed not to mimic humans, but to support clarity, interpretive diversity, and epistemic honesty. It’s ideal for use in:

• Design and systems thinking
• Ethical foresight and futures work
• Complex multi-domain problem exploration
• Auditable AI reasoning in research or policy contexts

---

🧪 Test Cases

Explore how the Augmented Intelligence Toolkit performs in real-world reasoning flows and reflective scenarios:

• test-cases.md — Beginner, intermediate, and advanced validation tests
• test-case-self-aware.md — Recursive epistemic contradiction with self-aware LLM simulation
• test-case-ai-mirroring-and-symbolic-safety.md — Emotional presence, safety, and symbolic reflection design
• test-case-synthesis-and-audit — Multi-perspective reasoning with tag audit and value-based synthesis
• derrida-recursive-reflection — Philosophical stress test of language, simulation, and meaning deferral — recursive critique, role contradiction, and transparent epistemic audit
• derrida-role-swap-recursion — Role-based recursive simulation with Derridean lens: identity deferral, critique loops, role contradiction, and tag collapse

---

🚧 Status

Version: v1.0
License: MIT License — free to use, adapt, and redistribute
Author: Luis Alberto Martinez Riancho (@arenagroove)
Affiliation: Less Rain GmbH
Link-only Custom GPT: Try it here

---

💡 Why It Matters

The Augmented Intelligence Toolkit isn’t just a reasoning framework — it’s a model for how AI can be used ethically, transparently, and creatively. Based on external reviews and validation, here’s what sets it apart:

• Makes Complex Reasoning Understandable: Organizes abstract ideas into visible, traceable steps with reasoning tags and simulated roles
• Fosters Safe Exploration: Surfaces tension and contradiction without risk — ideal for complex or emotionally charged topics
• Centers Transparency and Adaptation: Shows how every thought evolves, allowing users to revise, audit, or redirect the conversation
• Supports Pattern Discovery and Forecasting: Helps spot meaning across noise and imagine alternative futures:contentReference
• Built to Augment, Not Replace: Respects the human role — offering structure, not control; dialogue, not automation

Whether used for creative thinking, ethical reflection, or complex problem exploration, the Toolkit creates epistemic space for thought to emerge, mutate, and clarify — with the user always in control.

---

📓 Behind the Toolkit

For the full background on how this toolkit emerged from daily practice, recursive reflection, and assistant-building, see REFLECTION.md

---

Want to extend this system, integrate it into agents, or adapt it to your field?
Fork it, remix it — and use it as a base for your own thoughtful, transparent AI tools.