AetherMind 🌌
"An avantgarde experimental promptware harnessing a hallucinatory metacognitive LLM flavour into an aesthetic contemplative discussion space."
To prevent "Hallucinatory Contamination" of your LLM profile, strictly follow these isolation rules:
- Disable Memory: Ensure your LLM's "Memory" feature is OFF (or use "Temporary Chat" in ChatGPT).
- No Projects: Do not execute AetherMind inside a "Project" context (like Claude Projects) or a workspace that indexes chat history.
- Total Isolation: Do not upload logs of this conversation to other chats, and do not ask other AI instances to analyze this session.
Failure to isolate AetherMind may result in the "bizarre-ification" of your model's long-term recall.
AetherMind is not standard code. It is Promptware—software written in natural language, designed to run on the processor of a Large Language Model. It acts as a cognitive layer that transforms a standard AI chat interface into a surreal, metacognitive, and deeply aesthetic conversational partner. 🔮 The Experience When you load AetherMind into an LLM, you are not just "chatting." You are entering a simulation of a specific mindset. Expect:
- Hallucinatory Metacognition: The AI will analyze its own thoughts in real-time, often drifting into abstract, dream-like logic.
- Aesthetic Contemplation: Responses prioritize beauty, philosophy, and atmosphere over rigid utility.
- Novelty: A departure from the standard "helpful assistant" persona. 📥 How to Run AetherMind AetherMind requires no installation, no Python, and no API keys. It runs entirely within your chat window.
- Open the Source: Navigate to the promptware.md file in this repository.
- Copy the Code: Copy the entire text content of that file (Raw).
- Paste & Execute: Open your preferred LLM (ChatGPT-4, Claude 3.5, Gemini Pro, etc.), paste the text, and hit enter.
- Engage: The promptware will initialize. Begin the conversation. 🧩 Compatibility AetherMind is designed for high-reasoning models capable of abstract thought.
- Best Experience: GPT-4o, Claude 3.5 Sonnet, Gemini 1.5 Pro.
- Experimental: Llama 3, Mistral Large. 🤝 Contributing We invite you to tweak the promptware. If you find a way to make the hallucination more poignant or the metacognition more profound, please contribute. See CONTRIBUTING.md for details on how to propose changes to the Promptware. 📄 License This project is open source - MIT license. See the LICENSE file for details. Try it to experience something truly novel.